1 This is tar.info, produced by makeinfo version 4.13 from tar.texi.
3 This manual is for GNU `tar' (version 1.21, 30 October 2008), which
4 creates and extracts files from archives.
6 Copyright (C) 1992, 1994, 1995, 1996, 1997, 1999, 2000, 2001, 2003,
7 2004, 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
9 Permission is granted to copy, distribute and/or modify this
10 document under the terms of the GNU Free Documentation License,
11 Version 1.1 or any later version published by the Free Software
12 Foundation; with no Invariant Sections, with the Front-Cover Texts
13 being "A GNU Manual," and with the Back-Cover Texts as in (a)
14 below. A copy of the license is included in the section entitled
15 "GNU Free Documentation License".
17 (a) The FSF's Back-Cover Text is: "You have the freedom to copy
18 and modify this GNU manual. Buying copies from the FSF supports
19 it in developing GNU and promoting software freedom."
21 INFO-DIR-SECTION Archiving
23 * Tar: (tar). Making tape (or disk) archives.
26 INFO-DIR-SECTION Individual utilities
28 * tar: (tar)tar invocation. Invoking GNU `tar'.
32 File: tar.info, Node: gzip, Next: sparse, Up: Compression
34 8.1.1 Creating and Reading Compressed Archives
35 ----------------------------------------------
37 GNU `tar' is able to create and read compressed archives. It supports
38 `gzip', `bzip2', `lzma' and `lzop' compression programs. For backward
39 compatibility, it also supports `compress' command, although we
40 strongly recommend against using it, because it is by far less
41 effective than other compression programs(1).
43 Creating a compressed archive is simple: you just specify a
44 "compression option" along with the usual archive creation commands.
45 The compression option is `-z' (`--gzip') to create a `gzip' compressed
46 archive, `-j' (`--bzip2') to create a `bzip2' compressed archive, `-J'
47 (`--lzma') to create an LZMA compressed archive, `--lzop' to create an
48 LSOP archive, and `-Z' (`--compress') to use `compress' program. For
51 $ tar cfz archive.tar.gz .
53 You can also let GNU `tar' select the compression program basing on
54 the suffix of the archive file name. This is done using
55 `--auto-compress' (`-a') command line option. For example, the
56 following invocation will use `bzip2' for compression:
58 $ tar cfa archive.tar.bz2 .
60 whereas the following one will use `lzma':
62 $ tar cfa archive.tar.lzma .
64 For a complete list of file name suffixes recognized by GNU `tar',
65 *note auto-compress::.
67 Reading compressed archive is even simpler: you don't need to specify
68 any additional options as GNU `tar' recognizes its format
69 automatically. Thus, the following commands will list and extract the
70 archive created in previous example:
72 # List the compressed archive
73 $ tar tf archive.tar.gz
74 # Extract the compressed archive
75 $ tar xf archive.tar.gz
77 The format recognition algorithm is based on "signatures", a special
78 byte sequences in the beginning of file, that are specific for certain
79 compression formats. If this approach fails, `tar' falls back to using
80 archive name suffix to determine its format (*Note auto-compress::, for
81 a list of recognized suffixes).
83 The only case when you have to specify a decompression option while
84 reading the archive is when reading from a pipe or from a tape drive
85 that does not support random access. However, in this case GNU `tar'
86 will indicate which option you should use. For example:
88 $ cat archive.tar.gz | tar tf -
89 tar: Archive is compressed. Use -z option
90 tar: Error is not recoverable: exiting now
92 If you see such diagnostics, just add the suggested option to the
93 invocation of GNU `tar':
95 $ cat archive.tar.gz | tar tfz -
97 Notice also, that there are several restrictions on operations on
98 compressed archives. First of all, compressed archives cannot be
99 modified, i.e., you cannot update (`--update' (`-u')) them or delete
100 (`--delete') members from them or add (`--append' (`-r')) members to
101 them. Likewise, you cannot append another `tar' archive to a
102 compressed archive using `--concatenate' (`-A')). Secondly,
103 multi-volume archives cannot be compressed.
105 The following table summarizes compression options used by GNU `tar'.
109 Select a compression program to use by the archive file name
110 suffix. The following suffixes are recognized:
112 Suffix Compression program
113 --------------------------------------------------------------
130 Filter the archive through `gzip'.
132 You can use `--gzip' and `--gunzip' on physical devices (tape
133 drives, etc.) and remote files as well as on normal files; data to
134 or from such devices or remote files is reblocked by another copy
135 of the `tar' program to enforce the specified (or default) record
136 size. The default compression parameters are used; if you need to
137 override them, set `GZIP' environment variable, e.g.:
139 $ GZIP=--best tar cfz archive.tar.gz subdir
141 Another way would be to avoid the `--gzip' (`--gunzip',
142 `--ungzip', `-z') option and run `gzip' explicitly:
144 $ tar cf - subdir | gzip --best -c - > archive.tar.gz
146 About corrupted compressed archives: `gzip''ed files have no
147 redundancy, for maximum compression. The adaptive nature of the
148 compression scheme means that the compression tables are implicitly
149 spread all over the archive. If you lose a few blocks, the dynamic
150 construction of the compression tables becomes unsynchronized, and
151 there is little chance that you could recover later in the archive.
153 There are pending suggestions for having a per-volume or per-file
154 compression in GNU `tar'. This would allow for viewing the
155 contents without decompression, and for resynchronizing
156 decompression at every volume or file, in case of corrupted
157 archives. Doing so, we might lose some compressibility. But this
158 would have make recovering easier. So, there are pros and cons.
163 Filter the archive through `bzip2'. Otherwise like `--gzip'.
167 Filter the archive through `lzma'. Otherwise like `--gzip'.
170 Filter the archive through `lzop'. Otherwise like `--gzip'.
175 Filter the archive through `compress'. Otherwise like `--gzip'.
177 `--use-compress-program=PROG'
178 Use external compression program PROG. Use this option if you
179 have a compression program that GNU `tar' does not support. There
180 are two requirements to which PROG should comply:
182 First, when called without options, it should read data from
183 standard input, compress it and output it on standard output.
185 Secondly, if called with `-d' argument, it should do exactly the
186 opposite, i.e., read the compressed data from the standard input
187 and produce uncompressed data on the standard output.
189 The `--use-compress-program' option, in particular, lets you
190 implement your own filters, not necessarily dealing with
191 compression/decompression. For example, suppose you wish to implement
192 PGP encryption on top of compression, using `gpg' (*note gpg:
193 (gpg)Top.). The following script does that:
197 -d) gpg --decrypt - | gzip -d -c;;
198 '') gzip -c | gpg -s ;;
199 *) echo "Unknown option $1">&2; exit 1;;
202 Suppose you name it `gpgz' and save it somewhere in your `PATH'.
203 Then the following command will create a compressed archive signed with
206 $ tar -cf foo.tar.gpgz --use-compress=gpgz .
208 Likewise, the following command will list its contents:
210 $ tar -tf foo.tar.gpgz --use-compress=gpgz .
212 ---------- Footnotes ----------
214 (1) It also had patent problems in the past.
217 File: tar.info, Node: sparse, Prev: gzip, Up: Compression
219 8.1.2 Archiving Sparse Files
220 ----------------------------
222 Files in the file system occasionally have "holes". A "hole" in a file
223 is a section of the file's contents which was never written. The
224 contents of a hole reads as all zeros. On many operating systems,
225 actual disk storage is not allocated for holes, but they are counted in
226 the length of the file. If you archive such a file, `tar' could create
227 an archive longer than the original. To have `tar' attempt to
228 recognize the holes in a file, use `--sparse' (`-S'). When you use
229 this option, then, for any file using less disk space than would be
230 expected from its length, `tar' searches the file for consecutive
231 stretches of zeros. It then records in the archive for the file where
232 the consecutive stretches of zeros are, and only archives the "real
233 contents" of the file. On extraction (using `--sparse' is not needed
234 on extraction) any such files have holes created wherever the
235 continuous stretches of zeros were found. Thus, if you use `--sparse',
236 `tar' archives won't take more space than the original.
240 This option instructs `tar' to test each file for sparseness
241 before attempting to archive it. If the file is found to be
242 sparse it is treated specially, thus allowing to decrease the
243 amount of space used by its image in the archive.
245 This option is meaningful only when creating or updating archives.
246 It has no effect on extraction.
248 Consider using `--sparse' when performing file system backups, to
249 avoid archiving the expanded forms of files stored sparsely in the
252 Even if your system has no sparse files currently, some may be
253 created in the future. If you use `--sparse' while making file system
254 backups as a matter of course, you can be assured the archive will
255 never take more space on the media than the files take on disk
256 (otherwise, archiving a disk filled with sparse files might take
257 hundreds of tapes). *Note Incremental Dumps::.
259 However, be aware that `--sparse' option presents a serious
260 drawback. Namely, in order to determine if the file is sparse `tar'
261 has to read it before trying to archive it, so in total the file is
262 read *twice*. So, always bear in mind that the time needed to process
263 all files with this option is roughly twice the time needed to archive
266 When using `POSIX' archive format, GNU `tar' is able to store sparse
267 files using in three distinct ways, called "sparse formats". A sparse
268 format is identified by its "number", consisting, as usual of two
269 decimal numbers, delimited by a dot. By default, format `1.0' is used.
270 If, for some reason, you wish to use an earlier format, you can select
271 it using `--sparse-version' option.
273 `--sparse-version=VERSION'
274 Select the format to store sparse files in. Valid VERSION values
275 are: `0.0', `0.1' and `1.0'. *Note Sparse Formats::, for a
276 detailed description of each format.
278 Using `--sparse-format' option implies `--sparse'.
281 File: tar.info, Node: Attributes, Next: Portability, Prev: Compression, Up: Formats
283 8.2 Handling File Attributes
284 ============================
286 _(This message will disappear, once this node revised.)_
288 When `tar' reads files, it updates their access times. To avoid this,
289 use the `--atime-preserve[=METHOD]' option, which can either reset the
290 access time retroactively or avoid changing it in the first place.
292 Handling of file attributes
295 `--atime-preserve=replace'
296 `--atime-preserve=system'
297 Preserve the access times of files that are read. This works only
298 for files that you own, unless you have superuser privileges.
300 `--atime-preserve=replace' works on most systems, but it also
301 restores the data modification time and updates the status change
302 time. Hence it doesn't interact with incremental dumps nicely
303 (*note Incremental Dumps::), and it can set access or data
304 modification times incorrectly if other programs access the file
305 while `tar' is running.
307 `--atime-preserve=system' avoids changing the access time in the
308 first place, if the operating system supports this.
309 Unfortunately, this may or may not work on any given operating
310 system or file system. If `tar' knows for sure it won't work, it
311 complains right away.
313 Currently `--atime-preserve' with no operand defaults to
314 `--atime-preserve=replace', but this is intended to change to
315 `--atime-preserve=system' when the latter is better-supported.
319 Do not extract data modification time.
321 When this option is used, `tar' leaves the data modification times
322 of the files it extracts as the times when the files were
323 extracted, instead of setting it to the times recorded in the
326 This option is meaningless with `--list' (`-t').
329 Create extracted files with the same ownership they have in the
332 This is the default behavior for the superuser, so this option is
333 meaningful only for non-root users, when `tar' is executed on
334 those systems able to give files away. This is considered as a
335 security flaw by many people, at least because it makes quite
336 difficult to correctly account users for the disk space they
337 occupy. Also, the `suid' or `sgid' attributes of files are easily
338 and silently lost when files are given away.
340 When writing an archive, `tar' writes the user ID and user name
341 separately. If it can't find a user name (because the user ID is
342 not in `/etc/passwd'), then it does not write one. When restoring,
343 it tries to look the name (if one was written) up in
344 `/etc/passwd'. If it fails, then it uses the user ID stored in
349 Do not attempt to restore ownership when extracting. This is the
350 default behavior for ordinary users, so this option has an effect
351 only for the superuser.
354 The `--numeric-owner' option allows (ANSI) archives to be written
355 without user/group name information or such information to be
356 ignored when extracting. It effectively disables the generation
357 and/or use of user/group name information. This option forces
358 extraction using the numeric ids from the archive, ignoring the
361 This is useful in certain circumstances, when restoring a backup
362 from an emergency floppy with different passwd/group files for
363 example. It is otherwise impossible to extract files with the
364 right ownerships if the password file in use during the extraction
365 does not match the one belonging to the file system(s) being
366 extracted. This occurs, for example, if you are restoring your
367 files after a major crash and had booted from an emergency floppy
368 with no password file or put your disk into another machine to do
371 The numeric ids are _always_ saved into `tar' archives. The
372 identifying names are added at create time when provided by the
373 system, unless `--old-archive' (`-o') is used. Numeric ids could
374 be used when moving archives between a collection of machines using
375 a centralized management for attribution of numeric ids to users
376 and groups. This is often made through using the NIS capabilities.
378 When making a `tar' file for distribution to other sites, it is
379 sometimes cleaner to use a single owner for all files in the
380 distribution, and nicer to specify the write permission bits of the
381 files as stored in the archive independently of their actual value
382 on the file system. The way to prepare a clean distribution is
383 usually to have some Makefile rule creating a directory, copying
384 all needed files in that directory, then setting ownership and
385 permissions as wanted (there are a lot of possible schemes), and
386 only then making a `tar' archive out of this directory, before
387 cleaning everything out. Of course, we could add a lot of options
388 to GNU `tar' for fine tuning permissions and ownership. This is
389 not the good way, I think. GNU `tar' is already crowded with
390 options and moreover, the approach just explained gives you a
391 great deal of control already.
395 `--preserve-permissions'
396 Extract all protection information.
398 This option causes `tar' to set the modes (access permissions) of
399 extracted files exactly as recorded in the archive. If this option
400 is not used, the current `umask' setting limits the permissions on
401 extracted files. This option is by default enabled when `tar' is
402 executed by a superuser.
404 This option is meaningless with `--list' (`-t').
407 Same as both `--same-permissions' and `--same-order'.
409 The `--preserve' option has no equivalent short option name. It
410 is equivalent to `--same-permissions' plus `--same-order'.
414 File: tar.info, Node: Portability, Next: cpio, Prev: Attributes, Up: Formats
416 8.3 Making `tar' Archives More Portable
417 =======================================
419 Creating a `tar' archive on a particular system that is meant to be
420 useful later on many other machines and with other versions of `tar' is
421 more challenging than you might think. `tar' archive formats have been
422 evolving since the first versions of Unix. Many such formats are
423 around, and are not always compatible with each other. This section
424 discusses a few problems, and gives some advice about making `tar'
425 archives more portable.
427 One golden rule is simplicity. For example, limit your `tar'
428 archives to contain only regular files and directories, avoiding other
429 kind of special files. Do not attempt to save sparse files or
430 contiguous files as such. Let's discuss a few more problems, in turn.
434 * Portable Names:: Portable Names
435 * dereference:: Symbolic Links
436 * hard links:: Hard Links
437 * old:: Old V7 Archives
438 * ustar:: Ustar Archives
439 * gnu:: GNU and old GNU format archives.
440 * posix:: POSIX archives
441 * Checksumming:: Checksumming Problems
442 * Large or Negative Values:: Large files, negative time stamps, etc.
443 * Other Tars:: How to Extract GNU-Specific Data Using
444 Other `tar' Implementations
447 File: tar.info, Node: Portable Names, Next: dereference, Up: Portability
452 Use portable file and member names. A name is portable if it contains
453 only ASCII letters and digits, `/', `.', `_', and `-'; it cannot be
454 empty, start with `-' or `//', or contain `/-'. Avoid deep directory
455 nesting. For portability to old Unix hosts, limit your file name
456 components to 14 characters or less.
458 If you intend to have your `tar' archives to be read under MSDOS,
459 you should not rely on case distinction for file names, and you might
460 use the GNU `doschk' program for helping you further diagnosing illegal
461 MSDOS names, which are even more limited than System V's.
464 File: tar.info, Node: dereference, Next: hard links, Prev: Portable Names, Up: Portability
469 Normally, when `tar' archives a symbolic link, it writes a block to the
470 archive naming the target of the link. In that way, the `tar' archive
471 is a faithful record of the file system contents. `--dereference'
472 (`-h') is used with `--create' (`-c'), and causes `tar' to archive the
473 files symbolic links point to, instead of the links themselves. When
474 this option is used, when `tar' encounters a symbolic link, it will
475 archive the linked-to file, instead of simply recording the presence of
478 The name under which the file is stored in the file system is not
479 recorded in the archive. To record both the symbolic link name and the
480 file name in the system, archive the file under both names. If all
481 links were recorded automatically by `tar', an extracted file might be
482 linked to a file name that no longer exists in the file system.
484 If a linked-to file is encountered again by `tar' while creating the
485 same archive, an entire second copy of it will be stored. (This
486 _might_ be considered a bug.)
488 So, for portable archives, do not archive symbolic links as such,
489 and use `--dereference' (`-h'): many systems do not support symbolic
490 links, and moreover, your distribution might be unusable if it contains
491 unresolved symbolic links.
494 File: tar.info, Node: hard links, Next: old, Prev: dereference, Up: Portability
499 _(This message will disappear, once this node revised.)_
501 Normally, when `tar' archives a hard link, it writes a block to the
502 archive naming the target of the link (a `1' type block). In that way,
503 the actual file contents is stored in file only once. For example,
504 consider the following two files:
507 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 one
508 -rw-r--r-- 2 gray staff 4 2007-10-30 15:11 jeden
510 Here, `jeden' is a link to `one'. When archiving this directory
511 with a verbose level 2, you will get an output similar to the following:
513 $ tar cfvv ../archive.tar .
514 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
515 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
516 hrw-r--r-- gray/staff 0 2007-10-30 15:11 ./one link to ./jeden
518 The last line shows that, instead of storing two copies of the file,
519 `tar' stored it only once, under the name `jeden', and stored file
520 `one' as a hard link to this file.
522 It may be important to know that all hard links to the given file are
523 stored in the archive. For example, this may be necessary for exact
524 reproduction of the file system. The following option does that:
528 Check the number of links dumped for each processed file. If this
529 number does not match the total number of hard links for the file,
530 print a warning message.
532 For example, trying to archive only file `jeden' with this option
533 produces the following diagnostics:
535 $ tar -c -f ../archive.tar jeden
536 tar: Missing links to `jeden'.
538 Although creating special records for hard links helps keep a
539 faithful record of the file system contents and makes archives more
540 compact, it may present some difficulties when extracting individual
541 members from the archive. For example, trying to extract file `one'
542 from the archive created in previous examples produces, in the absense
545 $ tar xf archive.tar ./one
546 tar: ./one: Cannot hard link to `./jeden': No such file or directory
547 tar: Error exit delayed from previous errors
549 The reason for this behavior is that `tar' cannot seek back in the
550 archive to the previous member (in this case, `one'), to extract it(1).
551 If you wish to avoid such problems at the cost of a bigger archive, use
552 the following option:
555 Dereference hard links and store the files they refer to.
557 For example, trying this option on our two sample files, we get two
558 copies in the archive, each of which can then be extracted
559 independently of the other:
561 $ tar -c -vv -f ../archive.tar --hard-dereference .
562 drwxr-xr-x gray/staff 0 2007-10-30 15:13 ./
563 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./jeden
564 -rw-r--r-- gray/staff 4 2007-10-30 15:11 ./one
566 ---------- Footnotes ----------
568 (1) There are plans to fix this in future releases.
571 File: tar.info, Node: old, Next: ustar, Prev: hard links, Up: Portability
573 8.3.4 Old V7 Archives
574 ---------------------
576 Certain old versions of `tar' cannot handle additional information
577 recorded by newer `tar' programs. To create an archive in V7 format
578 (not ANSI), which can be read by these old versions, specify the
579 `--format=v7' option in conjunction with the `--create' (`-c') (`tar'
580 also accepts `--portability' or `--old-archive' for this option). When
581 you specify it, `tar' leaves out information about directories, pipes,
582 fifos, contiguous files, and device files, and specifies file ownership
583 by group and user IDs instead of group and user names.
585 When updating an archive, do not use `--format=v7' unless the
586 archive was created using this option.
588 In most cases, a _new_ format archive can be read by an _old_ `tar'
589 program without serious trouble, so this option should seldom be
590 needed. On the other hand, most modern `tar's are able to read old
591 format archives, so it might be safer for you to always use
592 `--format=v7' for your distributions. Notice, however, that `ustar'
593 format is a better alternative, as it is free from many of `v7''s
597 File: tar.info, Node: ustar, Next: gnu, Prev: old, Up: Portability
599 8.3.5 Ustar Archive Format
600 --------------------------
602 Archive format defined by POSIX.1-1988 specification is called `ustar'.
603 Although it is more flexible than the V7 format, it still has many
604 restrictions (*Note ustar: Formats, for the detailed description of
605 `ustar' format). Along with V7 format, `ustar' format is a good choice
606 for archives intended to be read with other implementations of `tar'.
608 To create archive in `ustar' format, use `--format=ustar' option in
609 conjunction with the `--create' (`-c').
612 File: tar.info, Node: gnu, Next: posix, Prev: ustar, Up: Portability
614 8.3.6 GNU and old GNU `tar' format
615 ----------------------------------
617 GNU `tar' was based on an early draft of the POSIX 1003.1 `ustar'
618 standard. GNU extensions to `tar', such as the support for file names
619 longer than 100 characters, use portions of the `tar' header record
620 which were specified in that POSIX draft as unused. Subsequent changes
621 in POSIX have allocated the same parts of the header record for other
622 purposes. As a result, GNU `tar' format is incompatible with the
623 current POSIX specification, and with `tar' programs that follow it.
625 In the majority of cases, `tar' will be configured to create this
626 format by default. This will change in future releases, since we plan
627 to make `POSIX' format the default.
629 To force creation a GNU `tar' archive, use option `--format=gnu'.
632 File: tar.info, Node: posix, Next: Checksumming, Prev: gnu, Up: Portability
634 8.3.7 GNU `tar' and POSIX `tar'
635 -------------------------------
637 Starting from version 1.14 GNU `tar' features full support for
638 POSIX.1-2001 archives.
640 A POSIX conformant archive will be created if `tar' was given
641 `--format=posix' (`--format=pax') option. No special option is
642 required to read and extract from a POSIX archive.
646 * PAX keywords:: Controlling Extended Header Keywords.
649 File: tar.info, Node: PAX keywords, Up: posix
651 8.3.7.1 Controlling Extended Header Keywords
652 ............................................
654 `--pax-option=KEYWORD-LIST'
655 Handle keywords in PAX extended headers. This option is
656 equivalent to `-o' option of the `pax' utility.
658 KEYWORD-LIST is a comma-separated list of keyword options, each
659 keyword option taking one of the following forms:
662 When used with one of archive-creation commands, this option
663 instructs `tar' to omit from extended header records that it
664 produces any keywords matching the string PATTERN.
666 When used in extract or list mode, this option instructs tar to
667 ignore any keywords matching the given PATTERN in the extended
668 header records. In both cases, matching is performed using the
669 pattern matching notation described in POSIX 1003.2, 3.13 (*note
670 wildcards::). For example:
672 --pax-option delete=security.*
674 would suppress security-related information.
677 This keyword allows user control over the name that is written
678 into the ustar header blocks for the extended headers. The name
679 is obtained from STRING after making the following substitutions:
681 Meta-character Replaced By
682 --------------------------------------------------------
683 %d The directory name of the file,
684 equivalent to the result of the
685 `dirname' utility on the translated
687 %f The name of the file with the
688 directory information stripped,
689 equivalent to the result of the
690 `basename' utility on the translated
692 %p The process ID of the `tar' process.
695 Any other `%' characters in STRING produce undefined results.
697 If no option `exthdr.name=string' is specified, `tar' will use the
698 following default value:
702 `globexthdr.name=STRING'
703 This keyword allows user control over the name that is written into
704 the ustar header blocks for global extended header records. The
705 name is obtained from the contents of STRING, after making the
706 following substitutions:
708 Meta-character Replaced By
709 --------------------------------------------------------
710 %n An integer that represents the
711 sequence number of the global
712 extended header record in the
713 archive, starting at 1.
714 %p The process ID of the `tar' process.
717 Any other `%' characters in STRING produce undefined results.
719 If no option `globexthdr.name=string' is specified, `tar' will use
720 the following default value:
722 $TMPDIR/GlobalHead.%p.%n
724 where `$TMPDIR' represents the value of the TMPDIR environment
725 variable. If TMPDIR is not set, `tar' uses `/tmp'.
728 When used with one of archive-creation commands, these
729 keyword/value pairs will be included at the beginning of the
730 archive in a global extended header record. When used with one of
731 archive-reading commands, `tar' will behave as if it has
732 encountered these keyword/value pairs at the beginning of the
733 archive in a global extended header record.
736 When used with one of archive-creation commands, these
737 keyword/value pairs will be included as records at the beginning
738 of an extended header for each file. This is effectively
739 equivalent to KEYWORD=VALUE form except that it creates no global
740 extended header records.
742 When used with one of archive-reading commands, `tar' will behave
743 as if these keyword/value pairs were included as records at the
744 end of each extended header; thus, they will override any global or
745 file-specific extended header record keywords of the same names.
746 For example, in the command:
748 tar --format=posix --create \
749 --file archive --pax-option gname:=user .
751 the group name will be forced to a new value for all files stored
755 File: tar.info, Node: Checksumming, Next: Large or Negative Values, Prev: posix, Up: Portability
757 8.3.8 Checksumming Problems
758 ---------------------------
760 SunOS and HP-UX `tar' fail to accept archives created using GNU `tar'
761 and containing non-ASCII file names, that is, file names having
762 characters with the eight bit set, because they use signed checksums,
763 while GNU `tar' uses unsigned checksums while creating archives, as per
764 POSIX standards. On reading, GNU `tar' computes both checksums and
765 accept any. It is somewhat worrying that a lot of people may go around
766 doing backup of their files using faulty (or at least non-standard)
767 software, not learning about it until it's time to restore their
768 missing files with an incompatible file extractor, or vice versa.
770 GNU `tar' compute checksums both ways, and accept any on read, so
771 GNU tar can read Sun tapes even with their wrong checksums. GNU `tar'
772 produces the standard checksum, however, raising incompatibilities with
773 Sun. That is to say, GNU `tar' has not been modified to _produce_
774 incorrect archives to be read by buggy `tar''s. I've been told that
775 more recent Sun `tar' now read standard archives, so maybe Sun did a
776 similar patch, after all?
778 The story seems to be that when Sun first imported `tar' sources on
779 their system, they recompiled it without realizing that the checksums
780 were computed differently, because of a change in the default signing
781 of `char''s in their compiler. So they started computing checksums
782 wrongly. When they later realized their mistake, they merely decided
783 to stay compatible with it, and with themselves afterwards.
784 Presumably, but I do not really know, HP-UX has chosen that their `tar'
785 archives to be compatible with Sun's. The current standards do not
786 favor Sun `tar' format. In any case, it now falls on the shoulders of
787 SunOS and HP-UX users to get a `tar' able to read the good archives
791 File: tar.info, Node: Large or Negative Values, Next: Other Tars, Prev: Checksumming, Up: Portability
793 8.3.9 Large or Negative Values
794 ------------------------------
796 _(This message will disappear, once this node revised.)_
798 The above sections suggest to use `oldest possible' archive format if
799 in doubt. However, sometimes it is not possible. If you attempt to
800 archive a file whose metadata cannot be represented using required
801 format, GNU `tar' will print error message and ignore such a file. You
802 will than have to switch to a format that is able to handle such
803 values. The format summary table (*note Formats::) will help you to do
806 In particular, when trying to archive files larger than 8GB or with
807 timestamps not in the range 1970-01-01 00:00:00 through 2242-03-16
808 12:56:31 UTC, you will have to chose between GNU and POSIX archive
809 formats. When considering which format to choose, bear in mind that
810 the GNU format uses two's-complement base-256 notation to store values
811 that do not fit into standard ustar range. Such archives can generally
812 be read only by a GNU `tar' implementation. Moreover, they sometimes
813 cannot be correctly restored on another hosts even by GNU `tar'. For
814 example, using two's complement representation for negative time stamps
815 that assumes a signed 32-bit `time_t' generates archives that are not
816 portable to hosts with differing `time_t' representations.
818 On the other hand, POSIX archives, generally speaking, can be
819 extracted by any tar implementation that understands older ustar
820 format. The only exception are files larger than 8GB.
823 File: tar.info, Node: Other Tars, Prev: Large or Negative Values, Up: Portability
825 8.3.10 How to Extract GNU-Specific Data Using Other `tar' Implementations
826 -------------------------------------------------------------------------
828 In previous sections you became acquainted with various quirks
829 necessary to make your archives portable. Sometimes you may need to
830 extract archives containing GNU-specific members using some third-party
831 `tar' implementation or an older version of GNU `tar'. Of course your
832 best bet is to have GNU `tar' installed, but if it is for some reason
833 impossible, this section will explain how to cope without it.
835 When we speak about "GNU-specific" members we mean two classes of
836 them: members split between the volumes of a multi-volume archive and
837 sparse members. You will be able to always recover such members if the
838 archive is in PAX format. In addition split members can be recovered
839 from archives in old GNU format. The following subsections describe
840 the required procedures in detail.
844 * Split Recovery:: Members Split Between Volumes
845 * Sparse Recovery:: Sparse Members
848 File: tar.info, Node: Split Recovery, Next: Sparse Recovery, Up: Other Tars
850 8.3.10.1 Extracting Members Split Between Volumes
851 .................................................
853 If a member is split between several volumes of an old GNU format
854 archive most third party `tar' implementation will fail to extract it.
855 To extract it, use `tarcat' program (*note Tarcat::). This program is
856 available from GNU `tar' home page
857 (http://www.gnu.org/software/tar/utils/tarcat.html). It concatenates
858 several archive volumes into a single valid archive. For example, if
859 you have three volumes named from `vol-1.tar' to `vol-3.tar', you can
860 do the following to extract them using a third-party `tar':
862 $ tarcat vol-1.tar vol-2.tar vol-3.tar | tar xf -
864 You could use this approach for most (although not all) PAX format
865 archives as well. However, extracting split members from a PAX archive
866 is a much easier task, because PAX volumes are constructed in such a
867 way that each part of a split member is extracted to a different file
868 by `tar' implementations that are not aware of GNU extensions. More
869 specifically, the very first part retains its original name, and all
870 subsequent parts are named using the pattern:
872 %d/GNUFileParts.%p/%f.%n
874 where symbols preceeded by `%' are "macro characters" that have the
877 Meta-character Replaced By
878 ------------------------------------------------------------
879 %d The directory name of the file,
880 equivalent to the result of the
881 `dirname' utility on its full name.
882 %f The file name of the file, equivalent
883 to the result of the `basename' utility
885 %p The process ID of the `tar' process that
887 %n Ordinal number of this particular part.
889 For example, if the file `var/longfile' was split during archive
890 creation between three volumes, and the creator `tar' process had
891 process ID `27962', then the member names will be:
894 var/GNUFileParts.27962/longfile.1
895 var/GNUFileParts.27962/longfile.2
897 When you extract your archive using a third-party `tar', these files
898 will be created on your disk, and the only thing you will need to do to
899 restore your file in its original form is concatenate them in the
900 proper order, for example:
903 $ cat GNUFileParts.27962/longfile.1 \
904 GNUFileParts.27962/longfile.2 >> longfile
905 $ rm -f GNUFileParts.27962
907 Notice, that if the `tar' implementation you use supports PAX format
908 archives, it will probably emit warnings about unknown keywords during
909 extraction. They will look like this:
912 Unknown extended header keyword 'GNU.volume.filename' ignored.
913 Unknown extended header keyword 'GNU.volume.size' ignored.
914 Unknown extended header keyword 'GNU.volume.offset' ignored.
916 You can safely ignore these warnings.
918 If your `tar' implementation is not PAX-aware, you will get more
919 warnings and more files generated on your disk, e.g.:
922 var/PaxHeaders.27962/longfile: Unknown file type 'x', extracted as
924 Unexpected EOF in archive
926 tmp/GlobalHead.27962.1: Unknown file type 'g', extracted as normal file
927 GNUFileParts.27962/PaxHeaders.27962/sparsefile.1: Unknown file type
928 'x', extracted as normal file
930 Ignore these warnings. The `PaxHeaders.*' directories created will
931 contain files with "extended header keywords" describing the extracted
932 files. You can delete them, unless they describe sparse members. Read
933 further to learn more about them.
936 File: tar.info, Node: Sparse Recovery, Prev: Split Recovery, Up: Other Tars
938 8.3.10.2 Extracting Sparse Members
939 ..................................
941 Any `tar' implementation will be able to extract sparse members from a
942 PAX archive. However, the extracted files will be "condensed", i.e.,
943 any zero blocks will be removed from them. When we restore such a
944 condensed file to its original form, by adding zero blocks (or "holes")
945 back to their original locations, we call this process "expanding" a
946 compressed sparse file.
948 To expand a file, you will need a simple auxiliary program called
949 `xsparse'. It is available in source form from GNU `tar' home page
950 (http://www.gnu.org/software/tar/utils/xsparse.html).
952 Let's begin with archive members in "sparse format version 1.0"(1),
953 which are the easiest to expand. The condensed file will contain both
954 file map and file data, so no additional data will be needed to restore
955 it. If the original file name was `DIR/NAME', then the condensed file
956 will be named `DIR/GNUSparseFile.N/NAME', where N is a decimal
959 To expand a version 1.0 file, run `xsparse' as follows:
961 $ xsparse `cond-file'
963 where `cond-file' is the name of the condensed file. The utility will
964 deduce the name for the resulting expanded file using the following
967 1. If `cond-file' does not contain any directories, `../cond-file'
970 2. If `cond-file' has the form `DIR/T/NAME', where both T and NAME
971 are simple names, with no `/' characters in them, the output file
972 name will be `DIR/NAME'.
974 3. Otherwise, if `cond-file' has the form `DIR/NAME', the output file
977 In the unlikely case when this algorithm does not suit your needs,
978 you can explicitly specify output file name as a second argument to the
981 $ xsparse `cond-file' `out-file'
983 It is often a good idea to run `xsparse' in "dry run" mode first.
984 In this mode, the command does not actually expand the file, but
985 verbosely lists all actions it would be taking to do so. The dry run
986 mode is enabled by `-n' command line argument:
988 $ xsparse -n /home/gray/GNUSparseFile.6058/sparsefile
989 Reading v.1.0 sparse map
990 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
991 `/home/gray/sparsefile'
994 To actually expand the file, you would run:
996 $ xsparse /home/gray/GNUSparseFile.6058/sparsefile
998 The program behaves the same way all UNIX utilities do: it will keep
999 quiet unless it has simething important to tell you (e.g. an error
1000 condition or something). If you wish it to produce verbose output,
1001 similar to that from the dry run mode, use `-v' option:
1003 $ xsparse -v /home/gray/GNUSparseFile.6058/sparsefile
1004 Reading v.1.0 sparse map
1005 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
1006 `/home/gray/sparsefile'
1009 Additionally, if your `tar' implementation has extracted the
1010 "extended headers" for this file, you can instruct `xstar' to use them
1011 in order to verify the integrity of the expanded file. The option `-x'
1012 sets the name of the extended header file to use. Continuing our
1015 $ xsparse -v -x /home/gray/PaxHeaders.6058/sparsefile \
1016 /home/gray/GNUSparseFile.6058/sparsefile
1017 Reading extended header file
1018 Found variable GNU.sparse.major = 1
1019 Found variable GNU.sparse.minor = 0
1020 Found variable GNU.sparse.name = sparsefile
1021 Found variable GNU.sparse.realsize = 217481216
1022 Reading v.1.0 sparse map
1023 Expanding file `/home/gray/GNUSparseFile.6058/sparsefile' to
1024 `/home/gray/sparsefile'
1027 An "extended header" is a special `tar' archive header that precedes
1028 an archive member and contains a set of "variables", describing the
1029 member properties that cannot be stored in the standard `ustar' header.
1030 While optional for expanding sparse version 1.0 members, the use of
1031 extended headers is mandatory when expanding sparse members in older
1032 sparse formats: v.0.0 and v.0.1 (The sparse formats are described in
1033 detail in *note Sparse Formats::.) So, for these formats, the question
1034 is: how to obtain extended headers from the archive?
1036 If you use a `tar' implementation that does not support PAX format,
1037 extended headers for each member will be extracted as a separate file.
1038 If we represent the member name as `DIR/NAME', then the extended header
1039 file will be named `DIR/PaxHeaders.N/NAME', where N is an integer
1042 Things become more difficult if your `tar' implementation does
1043 support PAX headers, because in this case you will have to manually
1044 extract the headers. We recommend the following algorithm:
1046 1. Consult the documentation of your `tar' implementation for an
1047 option that prints "block numbers" along with the archive listing
1048 (analogous to GNU `tar''s `-R' option). For example, `star' has
1051 2. Obtain verbose listing using the `block number' option, and find
1052 block numbers of the sparse member in question and the member
1053 immediately following it. For example, running `star' on our
1056 $ star -t -v -block-number -f arc.tar
1058 star: Unknown extended header keyword 'GNU.sparse.size' ignored.
1059 star: Unknown extended header keyword 'GNU.sparse.numblocks' ignored.
1060 star: Unknown extended header keyword 'GNU.sparse.name' ignored.
1061 star: Unknown extended header keyword 'GNU.sparse.map' ignored.
1062 block 56: 425984 -rw-r--r-- gray/users Jun 25 14:46 2006 GNUSparseFile.28124/sparsefile
1063 block 897: 65391 -rw-r--r-- gray/users Jun 24 20:06 2006 README
1066 (as usual, ignore the warnings about unknown keywords.)
1068 3. Let SIZE be the size of the sparse member, BS be its block number
1069 and BN be the block number of the next member. Compute:
1071 N = BS - BN - SIZE/512 - 2
1073 This number gives the size of the extended header part in tar
1074 "blocks". In our example, this formula gives: `897 - 56 - 425984
1077 4. Use `dd' to extract the headers:
1079 dd if=ARCHIVE of=HNAME bs=512 skip=BS count=N
1081 where ARCHIVE is the archive name, HNAME is a name of the file to
1082 store the extended header in, BS and N are computed in previous
1085 In our example, this command will be
1087 $ dd if=arc.tar of=xhdr bs=512 skip=56 count=7
1089 Finally, you can expand the condensed file, using the obtained
1092 $ xsparse -v -x xhdr GNUSparseFile.6058/sparsefile
1093 Reading extended header file
1094 Found variable GNU.sparse.size = 217481216
1095 Found variable GNU.sparse.numblocks = 208
1096 Found variable GNU.sparse.name = sparsefile
1097 Found variable GNU.sparse.map = 0,2048,1050624,2048,...
1098 Expanding file `GNUSparseFile.28124/sparsefile' to `sparsefile'
1101 ---------- Footnotes ----------
1105 (2) technically speaking, N is a "process ID" of the `tar' process
1106 which created the archive (*note PAX keywords::).
1109 File: tar.info, Node: cpio, Prev: Portability, Up: Formats
1111 8.4 Comparison of `tar' and `cpio'
1112 ==================================
1114 _(This message will disappear, once this node revised.)_
1116 The `cpio' archive formats, like `tar', do have maximum file name
1117 lengths. The binary and old ASCII formats have a maximum file length
1118 of 256, and the new ASCII and CRC ASCII formats have a max file length
1119 of 1024. GNU `cpio' can read and write archives with arbitrary file
1120 name lengths, but other `cpio' implementations may crash unexplainedly
1121 trying to read them.
1123 `tar' handles symbolic links in the form in which it comes in BSD;
1124 `cpio' doesn't handle symbolic links in the form in which it comes in
1125 System V prior to SVR4, and some vendors may have added symlinks to
1126 their system without enhancing `cpio' to know about them. Others may
1127 have enhanced it in a way other than the way I did it at Sun, and which
1128 was adopted by AT&T (and which is, I think, also present in the `cpio'
1129 that Berkeley picked up from AT&T and put into a later BSD release--I
1130 think I gave them my changes).
1132 (SVR4 does some funny stuff with `tar'; basically, its `cpio' can
1133 handle `tar' format input, and write it on output, and it probably
1134 handles symbolic links. They may not have bothered doing anything to
1135 enhance `tar' as a result.)
1137 `cpio' handles special files; traditional `tar' doesn't.
1139 `tar' comes with V7, System III, System V, and BSD source; `cpio'
1140 comes only with System III, System V, and later BSD (4.3-tahoe and
1143 `tar''s way of handling multiple hard links to a file can handle
1144 file systems that support 32-bit inumbers (e.g., the BSD file system);
1145 `cpio's way requires you to play some games (in its "binary" format,
1146 i-numbers are only 16 bits, and in its "portable ASCII" format, they're
1147 18 bits--it would have to play games with the "file system ID" field of
1148 the header to make sure that the file system ID/i-number pairs of
1149 different files were always different), and I don't know which `cpio's,
1150 if any, play those games. Those that don't might get confused and
1151 think two files are the same file when they're not, and make hard links
1154 `tar's way of handling multiple hard links to a file places only one
1155 copy of the link on the tape, but the name attached to that copy is the
1156 _only_ one you can use to retrieve the file; `cpio's way puts one copy
1157 for every link, but you can retrieve it using any of the names.
1159 What type of check sum (if any) is used, and how is this
1162 See the attached manual pages for `tar' and `cpio' format. `tar'
1163 uses a checksum which is the sum of all the bytes in the `tar' header
1164 for a file; `cpio' uses no checksum.
1166 If anyone knows why `cpio' was made when `tar' was present at the
1169 It wasn't. `cpio' first showed up in PWB/UNIX 1.0; no
1170 generally-available version of UNIX had `tar' at the time. I don't
1171 know whether any version that was generally available _within AT&T_ had
1172 `tar', or, if so, whether the people within AT&T who did `cpio' knew
1175 On restore, if there is a corruption on a tape `tar' will stop at
1176 that point, while `cpio' will skip over it and try to restore the rest
1179 The main difference is just in the command syntax and header format.
1181 `tar' is a little more tape-oriented in that everything is blocked
1182 to start on a record boundary.
1184 Is there any differences between the ability to recover crashed
1185 archives between the two of them. (Is there any chance of
1186 recovering crashed archives at all.)
1188 Theoretically it should be easier under `tar' since the blocking
1189 lets you find a header with some variation of `dd skip=NN'. However,
1190 modern `cpio''s and variations have an option to just search for the
1191 next file header after an error with a reasonable chance of resyncing.
1192 However, lots of tape driver software won't allow you to continue past
1193 a media error which should be the only reason for getting out of sync
1194 unless a file changed sizes while you were writing the archive.
1196 If anyone knows why `cpio' was made when `tar' was present at the
1197 unix scene, please tell me about this too.
1199 Probably because it is more media efficient (by not blocking
1200 everything and using only the space needed for the headers where `tar'
1201 always uses 512 bytes per file header) and it knows how to archive
1204 You might want to look at the freely available alternatives. The
1205 major ones are `afio', GNU `tar', and `pax', each of which have their
1206 own extensions with some backwards compatibility.
1208 Sparse files were `tar'red as sparse files (which you can easily
1209 test, because the resulting archive gets smaller, and GNU `cpio' can no
1213 File: tar.info, Node: Media, Next: Changes, Prev: Formats, Up: Top
1215 9 Tapes and Other Archive Media
1216 *******************************
1218 _(This message will disappear, once this node revised.)_
1220 A few special cases about tape handling warrant more detailed
1221 description. These special cases are discussed below.
1223 Many complexities surround the use of `tar' on tape drives. Since
1224 the creation and manipulation of archives located on magnetic tape was
1225 the original purpose of `tar', it contains many features making such
1226 manipulation easier.
1228 Archives are usually written on dismountable media--tape cartridges,
1229 mag tapes, or floppy disks.
1231 The amount of data a tape or disk holds depends not only on its size,
1232 but also on how it is formatted. A 2400 foot long reel of mag tape
1233 holds 40 megabytes of data when formatted at 1600 bits per inch. The
1234 physically smaller EXABYTE tape cartridge holds 2.3 gigabytes.
1236 Magnetic media are re-usable--once the archive on a tape is no longer
1237 needed, the archive can be erased and the tape or disk used over.
1238 Media quality does deteriorate with use, however. Most tapes or disks
1239 should be discarded when they begin to produce data errors. EXABYTE
1240 tape cartridges should be discarded when they generate an "error count"
1241 (number of non-usable bits) of more than 10k.
1243 Magnetic media are written and erased using magnetic fields, and
1244 should be protected from such fields to avoid damage to stored data.
1245 Sticking a floppy disk to a filing cabinet using a magnet is probably
1250 * Device:: Device selection and switching
1251 * Remote Tape Server::
1252 * Common Problems and Solutions::
1253 * Blocking:: Blocking
1254 * Many:: Many archives on one tape
1255 * Using Multiple Tapes:: Using Multiple Tapes
1256 * label:: Including a Label in the Archive
1258 * Write Protection::
1261 File: tar.info, Node: Device, Next: Remote Tape Server, Up: Media
1263 9.1 Device Selection and Switching
1264 ==================================
1266 _(This message will disappear, once this node revised.)_
1268 `-f [HOSTNAME:]FILE'
1269 `--file=[HOSTNAME:]FILE'
1270 Use archive file or device FILE on HOSTNAME.
1272 This option is used to specify the file name of the archive `tar'
1275 If the file name is `-', `tar' reads the archive from standard input
1276 (when listing or extracting), or writes it to standard output (when
1277 creating). If the `-' file name is given when updating an archive,
1278 `tar' will read the original archive from its standard input, and will
1279 write the entire new archive to its standard output.
1281 If the file name contains a `:', it is interpreted as `hostname:file
1282 name'. If the HOSTNAME contains an "at" sign (`@'), it is treated as
1283 `user@hostname:file name'. In either case, `tar' will invoke the
1284 command `rsh' (or `remsh') to start up an `/usr/libexec/rmt' on the
1285 remote machine. If you give an alternate login name, it will be given
1286 to the `rsh'. Naturally, the remote machine must have an executable
1287 `/usr/libexec/rmt'. This program is free software from the University
1288 of California, and a copy of the source code can be found with the
1289 sources for `tar'; it's compiled and installed by default. The exact
1290 path to this utility is determined when configuring the package. It is
1291 `PREFIX/libexec/rmt', where PREFIX stands for your installation prefix.
1292 This location may also be overridden at runtime by using
1293 `rmt-command=COMMAND' option (*Note --rmt-command: Option Summary, for
1294 detailed description of this option. *Note Remote Tape Server::, for
1295 the description of `rmt' command).
1297 If this option is not given, but the environment variable `TAPE' is
1298 set, its value is used; otherwise, old versions of `tar' used a default
1299 archive name (which was picked when `tar' was compiled). The default
1300 is normally set up to be the "first" tape drive or other transportable
1301 I/O medium on the system.
1303 Starting with version 1.11.5, GNU `tar' uses standard input and
1304 standard output as the default device, and I will not try anymore
1305 supporting automatic device detection at installation time. This was
1306 failing really in too many cases, it was hopeless. This is now
1307 completely left to the installer to override standard input and
1308 standard output for default device, if this seems preferable. Further,
1309 I think _most_ actual usages of `tar' are done with pipes or disks, not
1310 really tapes, cartridges or diskettes.
1312 Some users think that using standard input and output is running
1313 after trouble. This could lead to a nasty surprise on your screen if
1314 you forget to specify an output file name--especially if you are going
1315 through a network or terminal server capable of buffering large amounts
1316 of output. We had so many bug reports in that area of configuring
1317 default tapes automatically, and so many contradicting requests, that
1318 we finally consider the problem to be portably intractable. We could
1319 of course use something like `/dev/tape' as a default, but this is
1320 _also_ running after various kind of trouble, going from hung processes
1321 to accidental destruction of real tapes. After having seen all this
1322 mess, using standard input and output as a default really sounds like
1323 the only clean choice left, and a very useful one too.
1325 GNU `tar' reads and writes archive in records, I suspect this is the
1326 main reason why block devices are preferred over character devices.
1327 Most probably, block devices are more efficient too. The installer
1328 could also check for `DEFTAPE' in `<sys/mtio.h>'.
1331 Archive file is local even if it contains a colon.
1333 `--rsh-command=COMMAND'
1334 Use remote COMMAND instead of `rsh'. This option exists so that
1335 people who use something other than the standard `rsh' (e.g., a
1336 Kerberized `rsh') can access a remote device.
1338 When this command is not used, the shell command found when the
1339 `tar' program was installed is used instead. This is the first
1340 found of `/usr/ucb/rsh', `/usr/bin/remsh', `/usr/bin/rsh',
1341 `/usr/bsd/rsh' or `/usr/bin/nsh'. The installer may have
1342 overridden this by defining the environment variable `RSH' _at
1346 Specify drive and density.
1350 Create/list/extract multi-volume archive.
1352 This option causes `tar' to write a "multi-volume" archive--one
1353 that may be larger than will fit on the medium used to hold it.
1354 *Note Multi-Volume Archives::.
1358 Change tape after writing NUM x 1024 bytes.
1360 This option might be useful when your tape drivers do not properly
1361 detect end of physical tapes. By being slightly conservative on
1362 the maximum tape length, you might avoid the problem entirely.
1365 `--info-script=FILE'
1366 `--new-volume-script=FILE'
1367 Execute `file' at end of each tape. This implies `--multi-volume'
1368 (`-M'). *Note info-script::, for a detailed description of this
1372 File: tar.info, Node: Remote Tape Server, Next: Common Problems and Solutions, Prev: Device, Up: Media
1374 9.2 The Remote Tape Server
1375 ==========================
1377 In order to access the tape drive on a remote machine, `tar' uses the
1378 remote tape server written at the University of California at Berkeley.
1379 The remote tape server must be installed as `PREFIX/libexec/rmt' on any
1380 machine whose tape drive you want to use. `tar' calls `rmt' by running
1381 an `rsh' or `remsh' to the remote machine, optionally using a different
1382 login name if one is supplied.
1384 A copy of the source for the remote tape server is provided. It is
1385 Copyright (C) 1983 by the Regents of the University of California, but
1386 can be freely distributed. It is compiled and installed by default.
1388 Unless you use the `--absolute-names' (`-P') option, GNU `tar' will
1389 not allow you to create an archive that contains absolute file names (a
1390 file name beginning with `/'.) If you try, `tar' will automatically
1391 remove the leading `/' from the file names it stores in the archive.
1392 It will also type a warning message telling you what it is doing.
1394 When reading an archive that was created with a different `tar'
1395 program, GNU `tar' automatically extracts entries in the archive which
1396 have absolute file names as if the file names were not absolute. This
1397 is an important feature. A visitor here once gave a `tar' tape to an
1398 operator to restore; the operator used Sun `tar' instead of GNU `tar',
1399 and the result was that it replaced large portions of our `/bin' and
1400 friends with versions from the tape; needless to say, we were unhappy
1401 about having to recover the file system from backup tapes.
1403 For example, if the archive contained a file `/usr/bin/computoy',
1404 GNU `tar' would extract the file to `usr/bin/computoy', relative to the
1405 current directory. If you want to extract the files in an archive to
1406 the same absolute names that they had when the archive was created, you
1407 should do a `cd /' before extracting the files from the archive, or you
1408 should either use the `--absolute-names' option, or use the command
1411 Some versions of Unix (Ultrix 3.1 is known to have this problem),
1412 can claim that a short write near the end of a tape succeeded, when it
1413 actually failed. This will result in the -M option not working
1414 correctly. The best workaround at the moment is to use a significantly
1415 larger blocking factor than the default 20.
1417 In order to update an archive, `tar' must be able to backspace the
1418 archive in order to reread or rewrite a record that was just read (or
1419 written). This is currently possible only on two kinds of files: normal
1420 disk files (or any other file that can be backspaced with `lseek'), and
1421 industry-standard 9-track magnetic tape (or any other kind of tape that
1422 can be backspaced with the `MTIOCTOP' `ioctl'.
1424 This means that the `--append', `--concatenate', and `--delete'
1425 commands will not work on any other kind of file. Some media simply
1426 cannot be backspaced, which means these commands and options will never
1427 be able to work on them. These non-backspacing media include pipes and
1428 cartridge tape drives.
1430 Some other media can be backspaced, and `tar' will work on them once
1431 `tar' is modified to do so.
1433 Archives created with the `--multi-volume', `--label', and
1434 `--incremental' (`-G') options may not be readable by other version of
1435 `tar'. In particular, restoring a file that was split over a volume
1436 boundary will require some careful work with `dd', if it can be done at
1437 all. Other versions of `tar' may also create an empty file whose name
1438 is that of the volume header. Some versions of `tar' may create normal
1439 files instead of directories archived with the `--incremental' (`-G')
1443 File: tar.info, Node: Common Problems and Solutions, Next: Blocking, Prev: Remote Tape Server, Up: Media
1445 9.3 Some Common Problems and their Solutions
1446 ============================================
1450 no such file or directory
1454 directory checksum error
1457 errors from media/system:
1462 File: tar.info, Node: Blocking, Next: Many, Prev: Common Problems and Solutions, Up: Media
1467 _(This message will disappear, once this node revised.)_
1469 "Block" and "record" terminology is rather confused, and it is also
1470 confusing to the expert reader. On the other hand, readers who are new
1471 to the field have a fresh mind, and they may safely skip the next two
1472 paragraphs, as the remainder of this manual uses those two terms in a
1473 quite consistent way.
1475 John Gilmore, the writer of the public domain `tar' from which GNU
1476 `tar' was originally derived, wrote (June 1995):
1478 The nomenclature of tape drives comes from IBM, where I believe
1479 they were invented for the IBM 650 or so. On IBM mainframes, what
1480 is recorded on tape are tape blocks. The logical organization of
1481 data is into records. There are various ways of putting records
1482 into blocks, including `F' (fixed sized records), `V' (variable
1483 sized records), `FB' (fixed blocked: fixed size records, N to a
1484 block), `VB' (variable size records, N to a block), `VSB'
1485 (variable spanned blocked: variable sized records that can occupy
1486 more than one block), etc. The `JCL' `DD RECFORM=' parameter
1487 specified this to the operating system.
1489 The Unix man page on `tar' was totally confused about this. When
1490 I wrote `PD TAR', I used the historically correct terminology
1491 (`tar' writes data records, which are grouped into blocks). It
1492 appears that the bogus terminology made it into POSIX (no surprise
1493 here), and now Franc,ois has migrated that terminology back into
1494 the source code too.
1496 The term "physical block" means the basic transfer chunk from or to
1497 a device, after which reading or writing may stop without anything
1498 being lost. In this manual, the term "block" usually refers to a disk
1499 physical block, _assuming_ that each disk block is 512 bytes in length.
1500 It is true that some disk devices have different physical blocks, but
1501 `tar' ignore these differences in its own format, which is meant to be
1502 portable, so a `tar' block is always 512 bytes in length, and "block"
1503 always mean a `tar' block. The term "logical block" often represents
1504 the basic chunk of allocation of many disk blocks as a single entity,
1505 which the operating system treats somewhat atomically; this concept is
1506 only barely used in GNU `tar'.
1508 The term "physical record" is another way to speak of a physical
1509 block, those two terms are somewhat interchangeable. In this manual,
1510 the term "record" usually refers to a tape physical block, _assuming_
1511 that the `tar' archive is kept on magnetic tape. It is true that
1512 archives may be put on disk or used with pipes, but nevertheless, `tar'
1513 tries to read and write the archive one "record" at a time, whatever
1514 the medium in use. One record is made up of an integral number of
1515 blocks, and this operation of putting many disk blocks into a single
1516 tape block is called "reblocking", or more simply, "blocking". The
1517 term "logical record" refers to the logical organization of many
1518 characters into something meaningful to the application. The term
1519 "unit record" describes a small set of characters which are transmitted
1520 whole to or by the application, and often refers to a line of text.
1521 Those two last terms are unrelated to what we call a "record" in GNU
1524 When writing to tapes, `tar' writes the contents of the archive in
1525 chunks known as "records". To change the default blocking factor, use
1526 the `--blocking-factor=512-SIZE' (`-b 512-SIZE') option. Each record
1527 will then be composed of 512-SIZE blocks. (Each `tar' block is 512
1528 bytes. *Note Standard::.) Each file written to the archive uses at
1529 least one full record. As a result, using a larger record size can
1530 result in more wasted space for small files. On the other hand, a
1531 larger record size can often be read and written much more efficiently.
1533 Further complicating the problem is that some tape drives ignore the
1534 blocking entirely. For these, a larger record size can still improve
1535 performance (because the software layers above the tape drive still
1536 honor the blocking), but not as dramatically as on tape drives that
1539 When reading an archive, `tar' can usually figure out the record
1540 size on itself. When this is the case, and a non-standard record size
1541 was used when the archive was created, `tar' will print a message about
1542 a non-standard blocking factor, and then operate normally. On some
1543 tape devices, however, `tar' cannot figure out the record size itself.
1544 On most of those, you can specify a blocking factor (with
1545 `--blocking-factor') larger than the actual blocking factor, and then
1546 use the `--read-full-records' (`-B') option. (If you specify a
1547 blocking factor with `--blocking-factor' and don't use the
1548 `--read-full-records' option, then `tar' will not attempt to figure out
1549 the recording size itself.) On some devices, you must always specify
1550 the record size exactly with `--blocking-factor' when reading, because
1551 `tar' cannot figure it out. In any case, use `--list' (`-t') before
1552 doing any extractions to see whether `tar' is reading the archive
1555 `tar' blocks are all fixed size (512 bytes), and its scheme for
1556 putting them into records is to put a whole number of them (one or
1557 more) into each record. `tar' records are all the same size; at the
1558 end of the file there's a block containing all zeros, which is how you
1559 tell that the remainder of the last record(s) are garbage.
1561 In a standard `tar' file (no options), the block size is 512 and the
1562 record size is 10240, for a blocking factor of 20. What the
1563 `--blocking-factor' option does is sets the blocking factor, changing
1564 the record size while leaving the block size at 512 bytes. 20 was fine
1565 for ancient 800 or 1600 bpi reel-to-reel tape drives; most tape drives
1566 these days prefer much bigger records in order to stream and not waste
1567 tape. When writing tapes for myself, some tend to use a factor of the
1568 order of 2048, say, giving a record size of around one megabyte.
1570 If you use a blocking factor larger than 20, older `tar' programs
1571 might not be able to read the archive, so we recommend this as a limit
1572 to use in practice. GNU `tar', however, will support arbitrarily large
1573 record sizes, limited only by the amount of virtual memory or the
1574 physical characteristics of the tape device.
1578 * Format Variations:: Format Variations
1579 * Blocking Factor:: The Blocking Factor of an Archive
1582 File: tar.info, Node: Format Variations, Next: Blocking Factor, Up: Blocking
1584 9.4.1 Format Variations
1585 -----------------------
1587 _(This message will disappear, once this node revised.)_
1589 Format parameters specify how an archive is written on the archive
1590 media. The best choice of format parameters will vary depending on the
1591 type and number of files being archived, and on the media used to store
1594 To specify format parameters when accessing or creating an archive,
1595 you can use the options described in the following sections. If you do
1596 not specify any format parameters, `tar' uses default parameters. You
1597 cannot modify a compressed archive. If you create an archive with the
1598 `--blocking-factor' option specified (*note Blocking Factor::), you
1599 must specify that blocking-factor when operating on the archive. *Note
1600 Formats::, for other examples of format parameter considerations.
1603 File: tar.info, Node: Blocking Factor, Prev: Format Variations, Up: Blocking
1605 9.4.2 The Blocking Factor of an Archive
1606 ---------------------------------------
1608 _(This message will disappear, once this node revised.)_
1610 The data in an archive is grouped into blocks, which are 512 bytes.
1611 Blocks are read and written in whole number multiples called "records".
1612 The number of blocks in a record (i.e., the size of a record in units
1613 of 512 bytes) is called the "blocking factor". The
1614 `--blocking-factor=512-SIZE' (`-b 512-SIZE') option specifies the
1615 blocking factor of an archive. The default blocking factor is
1616 typically 20 (i.e., 10240 bytes), but can be specified at installation.
1617 To find out the blocking factor of an existing archive, use `tar --list
1618 --file=ARCHIVE-NAME'. This may not work on some devices.
1620 Records are separated by gaps, which waste space on the archive
1621 media. If you are archiving on magnetic tape, using a larger blocking
1622 factor (and therefore larger records) provides faster throughput and
1623 allows you to fit more data on a tape (because there are fewer gaps).
1624 If you are archiving on cartridge, a very large blocking factor (say
1625 126 or more) greatly increases performance. A smaller blocking factor,
1626 on the other hand, may be useful when archiving small files, to avoid
1627 archiving lots of nulls as `tar' fills out the archive to the end of
1628 the record. In general, the ideal record size depends on the size of
1629 the inter-record gaps on the tape you are using, and the average size
1630 of the files you are archiving. *Note create::, for information on
1633 Archives with blocking factors larger than 20 cannot be read by very
1634 old versions of `tar', or by some newer versions of `tar' running on
1635 old machines with small address spaces. With GNU `tar', the blocking
1636 factor of an archive is limited only by the maximum record size of the
1637 device containing the archive, or by the amount of available virtual
1640 Also, on some systems, not using adequate blocking factors, as
1641 sometimes imposed by the device drivers, may yield unexpected
1642 diagnostics. For example, this has been reported:
1644 Cannot write to /dev/dlt: Invalid argument
1646 In such cases, it sometimes happen that the `tar' bundled by the system
1647 is aware of block size idiosyncrasies, while GNU `tar' requires an
1648 explicit specification for the block size, which it cannot guess. This
1649 yields some people to consider GNU `tar' is misbehaving, because by
1650 comparison, `the bundle `tar' works OK'. Adding `-b 256', for example,
1651 might resolve the problem.
1653 If you use a non-default blocking factor when you create an archive,
1654 you must specify the same blocking factor when you modify that archive.
1655 Some archive devices will also require you to specify the blocking
1656 factor when reading that archive, however this is not typically the
1657 case. Usually, you can use `--list' (`-t') without specifying a
1658 blocking factor--`tar' reports a non-default record size and then lists
1659 the archive members as it would normally. To extract files from an
1660 archive with a non-standard blocking factor (particularly if you're not
1661 sure what the blocking factor is), you can usually use the
1662 `--read-full-records' (`-B') option while specifying a blocking factor
1663 larger then the blocking factor of the archive (i.e., `tar --extract
1664 --read-full-records --blocking-factor=300'. *Note list::, for more
1665 information on the `--list' (`-t') operation. *Note Reading::, for a
1666 more detailed explanation of that option.
1668 `--blocking-factor=NUMBER'
1670 Specifies the blocking factor of an archive. Can be used with any
1671 operation, but is usually not necessary with `--list' (`-t').
1676 `--blocking-factor=BLOCKS'
1677 Set record size to BLOCKS * 512 bytes.
1679 This option is used to specify a "blocking factor" for the archive.
1680 When reading or writing the archive, `tar', will do reads and
1681 writes of the archive in records of BLOCK*512 bytes. This is true
1682 even when the archive is compressed. Some devices requires that
1683 all write operations be a multiple of a certain size, and so, `tar'
1684 pads the archive out to the next record boundary.
1686 The default blocking factor is set when `tar' is compiled, and is
1687 typically 20. Blocking factors larger than 20 cannot be read by
1688 very old versions of `tar', or by some newer versions of `tar'
1689 running on old machines with small address spaces.
1691 With a magnetic tape, larger records give faster throughput and fit
1692 more data on a tape (because there are fewer inter-record gaps).
1693 If the archive is in a disk file or a pipe, you may want to specify
1694 a smaller blocking factor, since a large one will result in a large
1695 number of null bytes at the end of the archive.
1697 When writing cartridge or other streaming tapes, a much larger
1698 blocking factor (say 126 or more) will greatly increase
1699 performance. However, you must specify the same blocking factor
1700 when reading or updating the archive.
1702 Apparently, Exabyte drives have a physical block size of 8K bytes.
1703 If we choose our blocksize as a multiple of 8k bytes, then the
1704 problem seems to disappear. Id est, we are using block size of
1705 112 right now, and we haven't had the problem since we switched...
1707 With GNU `tar' the blocking factor is limited only by the maximum
1708 record size of the device containing the archive, or by the amount
1709 of available virtual memory.
1711 However, deblocking or reblocking is virtually avoided in a special
1712 case which often occurs in practice, but which requires all the
1713 following conditions to be simultaneously true:
1714 * the archive is subject to a compression option,
1716 * the archive is not handled through standard input or output,
1717 nor redirected nor piped,
1719 * the archive is directly handled to a local disk, instead of
1722 * `--blocking-factor' is not explicitly specified on the `tar'
1725 If the output goes directly to a local disk, and not through
1726 stdout, then the last write is not extended to a full record size.
1727 Otherwise, reblocking occurs. Here are a few other remarks on this
1730 * `gzip' will complain about trailing garbage if asked to
1731 uncompress a compressed archive on tape, there is an option
1732 to turn the message off, but it breaks the regularity of
1733 simply having to use `PROG -d' for decompression. It would
1734 be nice if gzip was silently ignoring any number of trailing
1735 zeros. I'll ask Jean-loup Gailly, by sending a copy of this
1738 * `compress' does not show this problem, but as Jean-loup
1739 pointed out to Michael, `compress -d' silently adds garbage
1740 after the result of decompression, which tar ignores because
1741 it already recognized its end-of-file indicator. So this bug
1742 may be safely ignored.
1744 * `gzip -d -q' will be silent about the trailing zeros indeed,
1745 but will still return an exit status of 2 which tar reports
1746 in turn. `tar' might ignore the exit status returned, but I
1747 hate doing that, as it weakens the protection `tar' offers
1748 users against other possible problems at decompression time.
1749 If `gzip' was silently skipping trailing zeros _and_ also
1750 avoiding setting the exit status in this innocuous case, that
1751 would solve this situation.
1753 * `tar' should become more solid at not stopping to read a pipe
1754 at the first null block encountered. This inelegantly breaks
1755 the pipe. `tar' should rather drain the pipe out before
1760 Ignore blocks of zeros in archive (means EOF).
1762 The `--ignore-zeros' (`-i') option causes `tar' to ignore blocks
1763 of zeros in the archive. Normally a block of zeros indicates the
1764 end of the archive, but when reading a damaged archive, or one
1765 which was created by concatenating several archives together, this
1766 option allows `tar' to read the entire archive. This option is
1767 not on by default because many versions of `tar' write garbage
1768 after the zeroed blocks.
1770 Note that this option causes `tar' to read to the end of the
1771 archive file, which may sometimes avoid problems when multiple
1772 files are stored on a single physical tape.
1775 `--read-full-records'
1776 Reblock as we read (for reading 4.2BSD pipes).
1778 If `--read-full-records' is used, `tar' will not panic if an
1779 attempt to read a record from the archive does not return a full
1780 record. Instead, `tar' will keep reading until it has obtained a
1783 This option is turned on by default when `tar' is reading an
1784 archive from standard input, or from a remote machine. This is
1785 because on BSD Unix systems, a read of a pipe will return however
1786 much happens to be in the pipe, even if it is less than `tar'
1787 requested. If this option was not used, `tar' would fail as soon
1788 as it read an incomplete record from the pipe.
1790 This option is also useful with the commands for updating an
1796 When handling various tapes or cartridges, you have to take care of
1797 selecting a proper blocking, that is, the number of disk blocks you put
1798 together as a single tape block on the tape, without intervening tape
1799 gaps. A "tape gap" is a small landing area on the tape with no
1800 information on it, used for decelerating the tape to a full stop, and
1801 for later regaining the reading or writing speed. When the tape driver
1802 starts reading a record, the record has to be read whole without
1803 stopping, as a tape gap is needed to stop the tape motion without
1804 loosing information.
1806 Using higher blocking (putting more disk blocks per tape block) will
1807 use the tape more efficiently as there will be less tape gaps. But
1808 reading such tapes may be more difficult for the system, as more memory
1809 will be required to receive at once the whole record. Further, if
1810 there is a reading error on a huge record, this is less likely that the
1811 system will succeed in recovering the information. So, blocking should
1812 not be too low, nor it should be too high. `tar' uses by default a
1813 blocking of 20 for historical reasons, and it does not really matter
1814 when reading or writing to disk. Current tape technology would easily
1815 accommodate higher blockings. Sun recommends a blocking of 126 for
1816 Exabytes and 96 for DATs. We were told that for some DLT drives, the
1817 blocking should be a multiple of 4Kb, preferably 64Kb (`-b 128') or 256
1818 for decent performance. Other manufacturers may use different
1819 recommendations for the same tapes. This might also depends of the
1820 buffering techniques used inside modern tape controllers. Some imposes
1821 a minimum blocking, or a maximum blocking. Others request blocking to
1822 be some exponent of two.
1824 So, there is no fixed rule for blocking. But blocking at read time
1825 should ideally be the same as blocking used at write time. At one place
1826 I know, with a wide variety of equipment, they found it best to use a
1827 blocking of 32 to guarantee that their tapes are fully interchangeable.
1829 I was also told that, for recycled tapes, prior erasure (by the same
1830 drive unit that will be used to create the archives) sometimes lowers
1831 the error rates observed at rewriting time.
1833 I might also use `--number-blocks' instead of `--block-number', so
1834 `--block' will then expand to `--blocking-factor' unambiguously.
1837 File: tar.info, Node: Many, Next: Using Multiple Tapes, Prev: Blocking, Up: Media
1839 9.5 Many Archives on One Tape
1840 =============================
1842 Most tape devices have two entries in the `/dev' directory, or entries
1843 that come in pairs, which differ only in the minor number for this
1844 device. Let's take for example `/dev/tape', which often points to the
1845 only or usual tape device of a given system. There might be a
1846 corresponding `/dev/nrtape' or `/dev/ntape'. The simpler name is the
1847 _rewinding_ version of the device, while the name having `nr' in it is
1848 the _no rewinding_ version of the same device.
1850 A rewinding tape device will bring back the tape to its beginning
1851 point automatically when this device is opened or closed. Since `tar'
1852 opens the archive file before using it and closes it afterwards, this
1853 means that a simple:
1855 $ tar cf /dev/tape DIRECTORY
1857 will reposition the tape to its beginning both prior and after saving
1858 DIRECTORY contents to it, thus erasing prior tape contents and making
1859 it so that any subsequent write operation will destroy what has just
1862 So, a rewinding device is normally meant to hold one and only one
1863 file. If you want to put more than one `tar' archive on a given tape,
1864 you will need to avoid using the rewinding version of the tape device.
1865 You will also have to pay special attention to tape positioning.
1866 Errors in positioning may overwrite the valuable data already on your
1867 tape. Many people, burnt by past experiences, will only use rewinding
1868 devices and limit themselves to one file per tape, precisely to avoid
1869 the risk of such errors. Be fully aware that writing at the wrong
1870 position on a tape loses all information past this point and most
1871 probably until the end of the tape, and this destroyed information
1872 _cannot_ be recovered.
1874 To save DIRECTORY-1 as a first archive at the beginning of a tape,
1875 and leave that tape ready for a second archive, you should use:
1877 $ mt -f /dev/nrtape rewind
1878 $ tar cf /dev/nrtape DIRECTORY-1
1880 "Tape marks" are special magnetic patterns written on the tape
1881 media, which are later recognizable by the reading hardware. These
1882 marks are used after each file, when there are many on a single tape.
1883 An empty file (that is to say, two tape marks in a row) signal the
1884 logical end of the tape, after which no file exist. Usually,
1885 non-rewinding tape device drivers will react to the close request issued
1886 by `tar' by first writing two tape marks after your archive, and by
1887 backspacing over one of these. So, if you remove the tape at that time
1888 from the tape drive, it is properly terminated. But if you write
1889 another file at the current position, the second tape mark will be
1890 erased by the new information, leaving only one tape mark between files.
1892 So, you may now save DIRECTORY-2 as a second archive after the first
1893 on the same tape by issuing the command:
1895 $ tar cf /dev/nrtape DIRECTORY-2
1897 and so on for all the archives you want to put on the same tape.
1899 Another usual case is that you do not write all the archives the same
1900 day, and you need to remove and store the tape between two archive
1901 sessions. In general, you must remember how many files are already
1902 saved on your tape. Suppose your tape already has 16 files on it, and
1903 that you are ready to write the 17th. You have to take care of skipping
1904 the first 16 tape marks before saving DIRECTORY-17, say, by using these
1907 $ mt -f /dev/nrtape rewind
1908 $ mt -f /dev/nrtape fsf 16
1909 $ tar cf /dev/nrtape DIRECTORY-17
1911 In all the previous examples, we put aside blocking considerations,
1912 but you should do the proper things for that as well. *Note Blocking::.
1916 * Tape Positioning:: Tape Positions and Tape Marks
1917 * mt:: The `mt' Utility
1920 File: tar.info, Node: Tape Positioning, Next: mt, Up: Many
1922 9.5.1 Tape Positions and Tape Marks
1923 -----------------------------------
1925 _(This message will disappear, once this node revised.)_
1927 Just as archives can store more than one file from the file system,
1928 tapes can store more than one archive file. To keep track of where
1929 archive files (or any other type of file stored on tape) begin and end,
1930 tape archive devices write magnetic "tape marks" on the archive media.
1931 Tape drives write one tape mark between files, two at the end of all
1934 If you think of data as a series of records "rrrr"'s, and tape marks
1935 as "*"'s, a tape might look like the following:
1937 rrrr*rrrrrr*rrrrr*rr*rrrrr**-------------------------
1939 Tape devices read and write tapes using a read/write "tape head"--a
1940 physical part of the device which can only access one point on the tape
1941 at a time. When you use `tar' to read or write archive data from a
1942 tape device, the device will begin reading or writing from wherever on
1943 the tape the tape head happens to be, regardless of which archive or
1944 what part of the archive the tape head is on. Before writing an
1945 archive, you should make sure that no data on the tape will be
1946 overwritten (unless it is no longer needed). Before reading an
1947 archive, you should make sure the tape head is at the beginning of the
1948 archive you want to read. You can do it manually via `mt' utility
1949 (*note mt::). The `restore' script does that automatically (*note
1950 Scripted Restoration::).
1952 If you want to add new archive file entries to a tape, you should
1953 advance the tape to the end of the existing file entries, backspace
1954 over the last tape mark, and write the new archive file. If you were
1955 to add two archives to the example above, the tape might look like the
1958 rrrr*rrrrrr*rrrrr*rr*rrrrr*rrr*rrrr**----------------
1961 File: tar.info, Node: mt, Prev: Tape Positioning, Up: Many
1963 9.5.2 The `mt' Utility
1964 ----------------------
1966 _(This message will disappear, once this node revised.)_
1968 *Note Blocking Factor::.
1970 You can use the `mt' utility to advance or rewind a tape past a
1971 specified number of archive files on the tape. This will allow you to
1972 move to the beginning of an archive before extracting or reading it, or
1973 to the end of all the archives before writing a new one.
1975 The syntax of the `mt' command is:
1977 mt [-f TAPENAME] OPERATION [NUMBER]
1979 where TAPENAME is the name of the tape device, NUMBER is the number
1980 of times an operation is performed (with a default of one), and
1981 OPERATION is one of the following:
1985 Writes NUMBER tape marks at the current position on the tape.
1988 Moves tape position forward NUMBER files.
1991 Moves tape position back NUMBER files.
1994 Rewinds the tape. (Ignores NUMBER).
1998 Rewinds the tape and takes the tape device off-line. (Ignores
2002 Prints status information about the tape unit.
2005 If you don't specify a TAPENAME, `mt' uses the environment variable
2006 `TAPE'; if `TAPE' is not set, `mt' will use the default device
2007 specified in your `sys/mtio.h' file (`DEFTAPE' variable). If this is
2008 not defined, the program will display a descriptive error message and
2011 `mt' returns a 0 exit status when the operation(s) were successful,
2012 1 if the command was unrecognized, and 2 if an operation failed.
2015 File: tar.info, Node: Using Multiple Tapes, Next: label, Prev: Many, Up: Media
2017 9.6 Using Multiple Tapes
2018 ========================
2020 Often you might want to write a large archive, one larger than will fit
2021 on the actual tape you are using. In such a case, you can run multiple
2022 `tar' commands, but this can be inconvenient, particularly if you are
2023 using options like `--exclude=PATTERN' or dumping entire file systems.
2024 Therefore, `tar' provides a special mode for creating multi-volume
2027 "Multi-volume" archive is a single `tar' archive, stored on several
2028 media volumes of fixed size. Although in this section we will often
2029 call `volume' a "tape", there is absolutely no requirement for
2030 multi-volume archives to be stored on tapes. Instead, they can use
2031 whatever media type the user finds convenient, they can even be located
2034 When creating a multi-volume archive, GNU `tar' continues to fill
2035 current volume until it runs out of space, then it switches to next
2036 volume (usually the operator is queried to replace the tape on this
2037 point), and continues working on the new volume. This operation
2038 continues until all requested files are dumped. If GNU `tar' detects
2039 end of media while dumping a file, such a file is archived in split
2040 form. Some very big files can even be split across several volumes.
2042 Each volume is itself a valid GNU `tar' archive, so it can be read
2043 without any special options. Consequently any file member residing
2044 entirely on one volume can be extracted or otherwise operated upon
2045 without needing the other volume. Sure enough, to extract a split
2046 member you would need all volumes its parts reside on.
2048 Multi-volume archives suffer from several limitations. In
2049 particular, they cannot be compressed.
2051 GNU `tar' is able to create multi-volume archives of two formats
2052 (*note Formats::): `GNU' and `POSIX'.
2056 * Multi-Volume Archives:: Archives Longer than One Tape or Disk
2057 * Tape Files:: Tape Files
2058 * Tarcat:: Concatenate Volumes into a Single Archive
2061 File: tar.info, Node: Multi-Volume Archives, Next: Tape Files, Up: Using Multiple Tapes
2063 9.6.1 Archives Longer than One Tape or Disk
2064 -------------------------------------------
2066 To create an archive that is larger than will fit on a single unit of
2067 the media, use the `--multi-volume' (`-M') option in conjunction with
2068 the `--create' option (*note create::). A "multi-volume" archive can
2069 be manipulated like any other archive (provided the `--multi-volume'
2070 option is specified), but is stored on more than one tape or disk.
2072 When you specify `--multi-volume', `tar' does not report an error
2073 when it comes to the end of an archive volume (when reading), or the
2074 end of the media (when writing). Instead, it prompts you to load a new
2075 storage volume. If the archive is on a magnetic tape, you should
2076 change tapes when you see the prompt; if the archive is on a floppy
2077 disk, you should change disks; etc.
2081 Creates a multi-volume archive, when used in conjunction with
2082 `--create' (`-c'). To perform any other operation on a
2083 multi-volume archive, specify `--multi-volume' in conjunction with
2084 that operation. For example:
2086 $ tar --create --multi-volume --file=/dev/tape FILES
2088 The method `tar' uses to detect end of tape is not perfect, and
2089 fails on some operating systems or on some devices. If `tar' cannot
2090 detect the end of the tape itself, you can use `--tape-length' option
2091 to inform it about the capacity of the tape:
2093 `--tape-length=SIZE'
2095 Set maximum length of a volume. The SIZE argument should then be
2096 the usable size of the tape in units of 1024 bytes. This option
2097 selects `--multi-volume' automatically. For example:
2099 $ tar --create --tape-length=41943040 --file=/dev/tape FILES
2101 When GNU `tar' comes to the end of a storage media, it asks you to
2102 change the volume. The built-in prompt for POSIX locale is(1):
2104 Prepare volume #N for `ARCHIVE' and hit return:
2106 where N is the ordinal number of the volume to be created and ARCHIVE
2107 is archive file or device name.
2109 When prompting for a new tape, `tar' accepts any of the following
2113 Request `tar' to explain possible responses
2116 Request `tar' to exit immediately.
2119 Request `tar' to write the next volume on the file FILE-NAME.
2122 Request `tar' to run a subshell. This option can be disabled by
2123 giving `--restrict' command line option to `tar'(2).
2126 Request `tar' to begin writing the next volume.
2128 (You should only type `y' after you have changed the tape; otherwise
2129 `tar' will write over the volume it just finished.)
2131 The volume number used by `tar' in its tape-changing prompt can be
2132 changed; if you give the `--volno-file=FILE-OF-NUMBER' option, then
2133 FILE-OF-NUMBER should be an non-existing file to be created, or else, a
2134 file already containing a decimal number. That number will be used as
2135 the volume number of the first volume written. When `tar' is finished,
2136 it will rewrite the file with the now-current volume number. (This does
2137 not change the volume number written on a tape label, as per *note
2138 label::, it _only_ affects the number used in the prompt.)
2140 If you want more elaborate behavior than this, you can write a
2141 special "new volume script", that will be responsible for changing the
2142 volume, and instruct `tar' to use it instead of its normal prompting
2145 `--info-script=SCRIPT-NAME'
2146 `--new-volume-script=SCRIPT-NAME'
2148 Specify the full name of the volume script to use. The script can
2149 be used to eject cassettes, or to broadcast messages such as
2150 `Someone please come change my tape' when performing unattended
2153 The SCRIPT-NAME is executed without any command line arguments. It
2154 inherits `tar''s shell environment. Additional data is passed to it
2155 via the following environment variables:
2158 GNU `tar' version number.
2161 The name of the archive `tar' is processing.
2163 `TAR_BLOCKING_FACTOR'
2164 Current blocking factor (*note Blocking::.
2167 Ordinal number of the volume `tar' is about to start.
2170 A short option describing the operation `tar' is executing *Note
2171 Operations::, for a complete list of subcommand options.
2174 Format of the archive being processed. *Note Formats::, for a
2175 complete list of archive format names.
2178 File descriptor which can be used to communicate the new volume
2181 The volume script can instruct `tar' to use new archive name, by
2182 writing in to file descriptor `$TAR_FD' (see below for an example).
2184 If the info script fails, `tar' exits; otherwise, it begins writing
2187 If you want `tar' to cycle through a series of files or tape drives,
2188 there are three approaches to choose from. First of all, you can give
2189 `tar' multiple `--file' options. In this case the specified files will
2190 be used, in sequence, as the successive volumes of the archive. Only
2191 when the first one in the sequence needs to be used again will `tar'
2192 prompt for a tape change (or run the info script). For example,
2193 suppose someone has two tape drives on a system named `/dev/tape0' and
2194 `/dev/tape1'. For having GNU `tar' to switch to the second drive when
2195 it needs to write the second tape, and then back to the first tape,
2196 etc., just do either of:
2198 $ tar --create --multi-volume --file=/dev/tape0 --file=/dev/tape1 FILES
2199 $ tar cMff /dev/tape0 /dev/tape1 FILES
2201 The second method is to use the `n' response to the tape-change
2204 Finally, the most flexible approach is to use a volume script, that
2205 writes new archive name to the file descriptor `$TAR_FD'. For example,
2206 the following volume script will create a series of archive files, named
2207 `ARCHIVE-VOL', where ARCHIVE is the name of the archive being created
2208 (as given by `--file' option) and VOL is the ordinal number of the
2209 archive being created:
2212 echo Preparing volume $TAR_VOLUME of $TAR_ARCHIVE.
2214 name=`expr $TAR_ARCHIVE : '\(.*\)-.*'`
2215 case $TAR_SUBCOMMAND in
2217 -d|-x|-t) test -r ${name:-$TAR_ARCHIVE}-$TAR_VOLUME || exit 1
2222 echo ${name:-$TAR_ARCHIVE}-$TAR_VOLUME >&$TAR_FD
2224 The same script can be used while listing, comparing or extracting
2225 from the created archive. For example:
2227 # Create a multi-volume archive:
2228 $ tar -c -L1024 -f archive.tar -F new-volume .
2229 # Extract from the created archive:
2230 $ tar -x -f archive.tar -F new-volume .
2232 Notice, that the first command had to use `-L' option, since otherwise
2233 GNU `tar' will end up writing everything to file `archive.tar'.
2235 You can read each individual volume of a multi-volume archive as if
2236 it were an archive by itself. For example, to list the contents of one
2237 volume, use `--list', without `--multi-volume' specified. To extract
2238 an archive member from one volume (assuming it is described that
2239 volume), use `--extract', again without `--multi-volume'.
2241 If an archive member is split across volumes (i.e., its entry begins
2242 on one volume of the media and ends on another), you need to specify
2243 `--multi-volume' to extract it successfully. In this case, you should
2244 load the volume where the archive member starts, and use `tar --extract
2245 --multi-volume'--`tar' will prompt for later volumes as it needs them.
2246 *Note extracting archives::, for more information about extracting
2249 Multi-volume archives can be modified like any other archive. To add
2250 files to a multi-volume archive, you need to only mount the last volume
2251 of the archive media (and new volumes, if needed). For all other
2252 operations, you need to use the entire archive.
2254 If a multi-volume archive was labeled using `--label=ARCHIVE-LABEL'
2255 (*note label::) when it was created, `tar' will not automatically label
2256 volumes which are added later. To label subsequent volumes, specify
2257 `--label=ARCHIVE-LABEL' again in conjunction with the `--append',
2258 `--update' or `--concatenate' operation.
2260 Notice that multi-volume support is a GNU extension and the archives
2261 created in this mode should be read only using GNU `tar'. If you
2262 absolutely have to process such archives using a third-party `tar'
2263 implementation, read *note Split Recovery::.
2265 ---------- Footnotes ----------
2267 (1) If you run GNU `tar' under a different locale, the translation
2268 to the locale's language will be used.
2270 (2) *Note --restrict::, for more information about this option
2273 File: tar.info, Node: Tape Files, Next: Tarcat, Prev: Multi-Volume Archives, Up: Using Multiple Tapes
2278 _(This message will disappear, once this node revised.)_
2280 To give the archive a name which will be recorded in it, use the
2281 `--label=VOLUME-LABEL' (`-V VOLUME-LABEL') option. This will write a
2282 special block identifying VOLUME-LABEL as the name of the archive to
2283 the front of the archive which will be displayed when the archive is
2284 listed with `--list'. If you are creating a multi-volume archive with
2285 `--multi-volume' (*note Using Multiple Tapes::), then the volume label
2286 will have `Volume NNN' appended to the name you give, where NNN is the
2287 number of the volume of the archive. (If you use the
2288 `--label=VOLUME-LABEL') option when reading an archive, it checks to
2289 make sure the label on the tape matches the one you give. *Note label::.
2291 When `tar' writes an archive to tape, it creates a single tape file.
2292 If multiple archives are written to the same tape, one after the other,
2293 they each get written as separate tape files. When extracting, it is
2294 necessary to position the tape at the right place before running `tar'.
2295 To do this, use the `mt' command. For more information on the `mt'
2296 command and on the organization of tapes into a sequence of tape files,
2299 People seem to often do:
2301 --label="SOME-PREFIX `date +SOME-FORMAT`"
2303 or such, for pushing a common date in all volumes or an archive set.
2306 File: tar.info, Node: Tarcat, Prev: Tape Files, Up: Using Multiple Tapes
2308 9.6.3 Concatenate Volumes into a Single Archive
2309 -----------------------------------------------
2311 Sometimes it is necessary to convert existing GNU `tar' multi-volume
2312 archive to a single `tar' archive. Simply concatenating all volumes
2313 into one will not work, since each volume carries an additional
2314 information at the beginning. GNU `tar' is shipped with the shell
2315 script `tarcat' designed for this purpose.
2317 The script takes a list of files comprising a multi-volume archive
2318 and creates the resulting archive at the standard output. For example:
2320 tarcat vol.1 vol.2 vol.3 | tar tf -
2322 The script implements a simple heuristics to determine the format of
2323 the first volume file and to decide how to process the rest of the
2324 files. However, it makes no attempt to verify whether the files are
2325 given in order or even if they are valid `tar' archives. It uses `dd'
2326 and does not filter its standard error, so you will usually see lots of
2330 File: tar.info, Node: label, Next: verify, Prev: Using Multiple Tapes, Up: Media
2332 9.7 Including a Label in the Archive
2333 ====================================
2335 _(This message will disappear, once this node revised.)_
2337 To avoid problems caused by misplaced paper labels on the archive
2338 media, you can include a "label" entry--an archive member which
2339 contains the name of the archive--in the archive itself. Use the
2340 `--label=ARCHIVE-LABEL' (`-V ARCHIVE-LABEL') option in conjunction with
2341 the `--create' operation to include a label entry in the archive as it
2344 `--label=ARCHIVE-LABEL'
2346 Includes an "archive-label" at the beginning of the archive when
2347 the archive is being created, when used in conjunction with the
2348 `--create' operation. Checks to make sure the archive label
2349 matches the one specified (when used in conjunction with any other
2352 If you create an archive using both `--label=ARCHIVE-LABEL' (`-V
2353 ARCHIVE-LABEL') and `--multi-volume' (`-M'), each volume of the archive
2354 will have an archive label of the form `ARCHIVE-LABEL Volume N', where
2355 N is 1 for the first volume, 2 for the next, and so on. *Note Using
2356 Multiple Tapes::, for information on creating multiple volume archives.
2358 The volume label will be displayed by `--list' along with the file
2359 contents. If verbose display is requested, it will also be explicitly
2360 marked as in the example below:
2362 $ tar --verbose --list --file=iamanarchive
2363 V--------- 0 0 0 1992-03-07 12:01 iamalabel--Volume Header--
2364 -rw-r--r-- ringo user 40 1990-05-21 13:30 iamafilename
2366 However, `--list' option will cause listing entire contents of the
2367 archive, which may be undesirable (for example, if the archive is
2368 stored on a tape). You can request checking only the volume by
2369 specifying `--test-label' option. This option reads only the first
2370 block of an archive, so it can be used with slow storage devices. For
2373 $ tar --test-label --file=iamanarchive
2376 If `--test-label' is used with a single command line argument, `tar'
2377 compares the volume label with the argument. It exits with code 0 if
2378 the two strings match, and with code 2 otherwise. In this case no
2379 output is displayed. For example:
2381 $ tar --test-label --file=iamanarchive 'iamalable'
2383 $ tar --test-label --file=iamanarchive 'iamalable' alabel
2386 If you request any operation, other than `--create', along with
2387 using `--label' option, `tar' will first check if the archive label
2388 matches the one specified and will refuse to proceed if it does not.
2389 Use this as a safety precaution to avoid accidentally overwriting
2390 existing archives. For example, if you wish to add files to `archive',
2391 presumably labeled with string `My volume', you will get:
2393 $ tar -rf archive --label 'My volume' .
2394 tar: Archive not labeled to match `My volume'
2396 in case its label does not match. This will work even if `archive' is
2399 Similarly, `tar' will refuse to list or extract the archive if its
2400 label doesn't match the ARCHIVE-LABEL specified. In those cases,
2401 ARCHIVE-LABEL argument is interpreted as a globbing-style pattern which
2402 must match the actual magnetic volume label. *Note exclude::, for a
2403 precise description of how match is attempted(1). If the switch
2404 `--multi-volume' (`-M') is being used, the volume label matcher will
2405 also suffix ARCHIVE-LABEL by ` Volume [1-9]*' if the initial match
2406 fails, before giving up. Since the volume numbering is automatically
2407 added in labels at creation time, it sounded logical to equally help
2408 the user taking care of it when the archive is being read.
2410 The `--label' was once called `--volume', but is not available under
2413 You can also use `--label' to get a common information on all tapes
2414 of a series. For having this information different in each series
2415 created through a single script used on a regular basis, just manage to
2416 get some date string as part of the label. For example:
2418 $ tar cfMV /dev/tape "Daily backup for `date +%Y-%m-%d`"
2419 $ tar --create --file=/dev/tape --multi-volume \
2420 --volume="Daily backup for `date +%Y-%m-%d`"
2422 Also note that each label has its own date and time, which
2423 corresponds to when GNU `tar' initially attempted to write it, often
2424 soon after the operator launches `tar' or types the carriage return
2425 telling that the next tape is ready. Comparing date labels does give
2426 an idea of tape throughput only if the delays for rewinding tapes and
2427 the operator switching them were negligible, which is usually not the
2430 ---------- Footnotes ----------
2432 (1) Previous versions of `tar' used full regular expression
2433 matching, or before that, only exact string matching, instead of
2434 wildcard matchers. We decided for the sake of simplicity to use a
2435 uniform matching device through `tar'.
2438 File: tar.info, Node: verify, Next: Write Protection, Prev: label, Up: Media
2440 9.8 Verifying Data as It is Stored
2441 ==================================
2445 Attempt to verify the archive after writing.
2447 This option causes `tar' to verify the archive after writing it.
2448 Each volume is checked after it is written, and any discrepancies are
2449 recorded on the standard error output.
2451 Verification requires that the archive be on a back-space-able
2452 medium. This means pipes, some cartridge tape drives, and some other
2453 devices cannot be verified.
2455 You can insure the accuracy of an archive by comparing files in the
2456 system with archive members. `tar' can compare an archive to the file
2457 system as the archive is being written, to verify a write operation, or
2458 can compare a previously written archive, to insure that it is up to
2461 To check for discrepancies in an archive immediately after it is
2462 written, use the `--verify' (`-W') option in conjunction with the
2463 `--create' operation. When this option is specified, `tar' checks
2464 archive members against their counterparts in the file system, and
2465 reports discrepancies on the standard error.
2467 To verify an archive, you must be able to read it from before the end
2468 of the last written entry. This option is useful for detecting data
2469 errors on some tapes. Archives written to pipes, some cartridge tape
2470 drives, and some other devices cannot be verified.
2472 One can explicitly compare an already made archive with the file
2473 system by using the `--compare' (`--diff', `-d') option, instead of
2474 using the more automatic `--verify' option. *Note compare::.
2476 Note that these two options have a slightly different intent. The
2477 `--compare' option checks how identical are the logical contents of some
2478 archive with what is on your disks, while the `--verify' option is
2479 really for checking if the physical contents agree and if the recording
2480 media itself is of dependable quality. So, for the `--verify'
2481 operation, `tar' tries to defeat all in-memory cache pertaining to the
2482 archive, while it lets the speed optimization undisturbed for the
2483 `--compare' option. If you nevertheless use `--compare' for media
2484 verification, you may have to defeat the in-memory cache yourself,
2485 maybe by opening and reclosing the door latch of your recording unit,
2486 forcing some doubt in your operating system about the fact this is
2487 really the same volume as the one just written or read.
2489 The `--verify' option would not be necessary if drivers were indeed
2490 able to detect dependably all write failures. This sometimes require
2491 many magnetic heads, some able to read after the writes occurred. One
2492 would not say that drivers unable to detect all cases are necessarily
2493 flawed, as long as programming is concerned.
2495 The `--verify' (`-W') option will not work in conjunction with the
2496 `--multi-volume' (`-M') option or the `--append' (`-r'), `--update'
2497 (`-u') and `--delete' operations. *Note Operations::, for more
2498 information on these operations.
2500 Also, since `tar' normally strips leading `/' from file names (*note
2501 absolute::), a command like `tar --verify -cf /tmp/foo.tar /etc' will
2502 work as desired only if the working directory is `/', as `tar' uses the
2503 archive's relative member names (e.g., `etc/motd') when verifying the
2507 File: tar.info, Node: Write Protection, Prev: verify, Up: Media
2509 9.9 Write Protection
2510 ====================
2512 Almost all tapes and diskettes, and in a few rare cases, even disks can
2513 be "write protected", to protect data on them from being changed. Once
2514 an archive is written, you should write protect the media to prevent
2515 the archive from being accidentally overwritten or deleted. (This will
2516 protect the archive from being changed with a tape or floppy drive--it
2517 will not protect it from magnet fields or other physical hazards).
2519 The write protection device itself is usually an integral part of the
2520 physical media, and can be a two position (write enabled/write
2521 disabled) switch, a notch which can be popped out or covered, a ring
2522 which can be removed from the center of a tape reel, or some other
2526 File: tar.info, Node: Changes, Next: Configuring Help Summary, Prev: Media, Up: Top
2531 This appendix lists some important user-visible changes between version
2532 GNU `tar' 1.21 and previous versions. An up-to-date version of this
2533 document is available at the GNU `tar' documentation page
2534 (http://www.gnu.org/software/tar/manual/changes.html).
2536 Use of globbing patterns when listing and extracting.
2537 Previous versions of GNU tar assumed shell-style globbing when
2538 extracting from or listing an archive. For example:
2540 $ tar xf foo.tar '*.c'
2542 would extract all files whose names end in `.c'. This behavior
2543 was not documented and was incompatible with traditional tar
2544 implementations. Therefore, starting from version 1.15.91, GNU tar
2545 no longer uses globbing by default. For example, the above
2546 invocation is now interpreted as a request to extract from the
2547 archive the file named `*.c'.
2549 To facilitate transition to the new behavior for those users who
2550 got used to the previous incorrect one, `tar' will print a warning
2551 if it finds out that a requested member was not found in the
2552 archive and its name looks like a globbing pattern. For example:
2554 $ tar xf foo.tar '*.c'
2555 tar: Pattern matching characters used in file names. Please,
2556 tar: use --wildcards to enable pattern matching, or --no-wildcards to
2557 tar: suppress this warning.
2558 tar: *.c: Not found in archive
2559 tar: Error exit delayed from previous errors
2561 To treat member names as globbing patterns, use -wildcards option.
2562 If you want to tar to mimic the behavior of versions prior to
2563 1.15.91, add this option to your `TAR_OPTIONS' variable.
2565 *Note wildcards::, for the detailed discussion of the use of
2566 globbing patterns by GNU `tar'.
2568 Use of short option `-o'.
2569 Earlier versions of GNU `tar' understood `-o' command line option
2570 as a synonym for `--old-archive'.
2572 GNU `tar' starting from version 1.13.90 understands this option as
2573 a synonym for `--no-same-owner'. This is compatible with UNIX98
2574 `tar' implementations.
2576 However, to facilitate transition, `-o' option retains its old
2577 semantics when it is used with one of archive-creation commands.
2578 Users are encouraged to use `--format=oldgnu' instead.
2580 It is especially important, since versions of GNU Automake up to
2581 and including 1.8.4 invoke tar with this option to produce
2582 distribution tarballs. *Note v7: Formats, for the detailed
2583 discussion of this issue and its implications.
2585 . *Note tar-v7: (automake)Options, for a description on how to
2586 use various archive formats with `automake'.
2588 Future versions of GNU `tar' will understand `-o' only as a
2589 synonym for `--no-same-owner'.
2591 Use of short option `-l'
2592 Earlier versions of GNU `tar' understood `-l' option as a synonym
2593 for `--one-file-system'. Since such usage contradicted to UNIX98
2594 specification and harmed compatibility with other implementation,
2595 it was declared deprecated in version 1.14. However, to
2596 facilitate transition to its new semantics, it was supported by
2597 versions 1.15 and 1.15.90. The present use of `-l' as a short
2598 variant of `--check-links' was introduced in version 1.15.91.
2600 Use of options `--portability' and `--old-archive'
2601 These options are deprecated. Please use `--format=v7' instead.
2603 Use of option `--posix'
2604 This option is deprecated. Please use `--format=posix' instead.
2607 File: tar.info, Node: Configuring Help Summary, Next: Fixing Snapshot Files, Prev: Changes, Up: Top
2609 Appendix B Configuring Help Summary
2610 ***********************************
2612 Running `tar --help' displays the short `tar' option summary (*note
2613 help::). This summary is organized by "groups" of semantically close
2614 options. The options within each group are printed in the following
2615 order: a short option, eventually followed by a list of corresponding
2616 long option names, followed by a short description of the option. For
2617 example, here is an excerpt from the actual `tar --help' output:
2619 Main operation mode:
2621 -A, --catenate, --concatenate append tar files to an archive
2622 -c, --create create a new archive
2623 -d, --diff, --compare find differences between archive and
2625 --delete delete from the archive
2627 The exact visual representation of the help output is configurable
2628 via `ARGP_HELP_FMT' environment variable. The value of this variable is
2629 a comma-separated list of "format variable" assignments. There are two
2630 kinds of format variables. An "offset variable" keeps the offset of
2631 some part of help output text from the leftmost column on the screen. A
2632 "boolean" variable is a flag that toggles some output feature on or
2633 off. Depending on the type of the corresponding variable, there are two
2634 kinds of assignments:
2637 The assignment to an offset variable has the following syntax:
2641 where VARIABLE is the variable name, and VALUE is a numeric value
2642 to be assigned to the variable.
2645 To assign `true' value to a variable, simply put this variable
2646 name. To assign `false' value, prefix the variable name with
2649 # Assign `true' value:
2651 # Assign `false' value:
2654 Following variables are declared:
2656 -- Help Output: boolean dup-args
2657 If true, arguments for an option are shown with both short and long
2658 options, even when a given option has both forms, for example:
2660 -f ARCHIVE, --file=ARCHIVE use archive file or device ARCHIVE
2662 If false, then if an option has both short and long forms, the
2663 argument is only shown with the long one, for example:
2665 -f, --file=ARCHIVE use archive file or device ARCHIVE
2667 and a message indicating that the argument is applicable to both
2668 forms is printed below the options. This message can be disabled
2669 using `dup-args-note' (see below).
2671 The default is false.
2673 -- Help Output: boolean dup-args-note
2674 If this variable is true, which is the default, the following
2675 notice is displayed at the end of the help output:
2677 Mandatory or optional arguments to long options are also
2678 mandatory or optional for any corresponding short options.
2680 Setting `no-dup-args-note' inhibits this message. Normally, only
2681 one of variables `dup-args' or `dup-args-note' should be set.
2683 -- Help Output: offset short-opt-col
2684 Column in which short options start. Default is 2.
2686 $ tar --help|grep ARCHIVE
2687 -f, --file=ARCHIVE use archive file or device ARCHIVE
2688 $ ARGP_HELP_FMT=short-opt-col=6 tar --help|grep ARCHIVE
2689 -f, --file=ARCHIVE use archive file or device ARCHIVE
2691 -- Help Output: offset long-opt-col
2692 Column in which long options start. Default is 6. For example:
2694 $ tar --help|grep ARCHIVE
2695 -f, --file=ARCHIVE use archive file or device ARCHIVE
2696 $ ARGP_HELP_FMT=long-opt-col=16 tar --help|grep ARCHIVE
2697 -f, --file=ARCHIVE use archive file or device ARCHIVE
2699 -- Help Output: offset doc-opt-col
2700 Column in which "doc options" start. A doc option isn't actually
2701 an option, but rather an arbitrary piece of documentation that is
2702 displayed in much the same manner as the options. For example, in
2703 the description of `--format' option:
2705 -H, --format=FORMAT create archive of the given format.
2707 FORMAT is one of the following:
2709 gnu GNU tar 1.13.x format
2710 oldgnu GNU format as per tar <= 1.12
2711 pax POSIX 1003.1-2001 (pax) format
2713 ustar POSIX 1003.1-1988 (ustar) format
2714 v7 old V7 tar format
2716 the format names are doc options. Thus, if you set
2717 `ARGP_HELP_FMT=doc-opt-col=6' the above part of the help output
2718 will look as follows:
2720 -H, --format=FORMAT create archive of the given format.
2722 FORMAT is one of the following:
2724 gnu GNU tar 1.13.x format
2725 oldgnu GNU format as per tar <= 1.12
2726 pax POSIX 1003.1-2001 (pax) format
2728 ustar POSIX 1003.1-1988 (ustar) format
2729 v7 old V7 tar format
2731 -- Help Output: offset opt-doc-col
2732 Column in which option description starts. Default is 29.
2734 $ tar --help|grep ARCHIVE
2735 -f, --file=ARCHIVE use archive file or device ARCHIVE
2736 $ ARGP_HELP_FMT=opt-doc-col=19 tar --help|grep ARCHIVE
2737 -f, --file=ARCHIVE use archive file or device ARCHIVE
2738 $ ARGP_HELP_FMT=opt-doc-col=9 tar --help|grep ARCHIVE
2740 use archive file or device ARCHIVE
2742 Notice, that the description starts on a separate line if
2743 `opt-doc-col' value is too small.
2745 -- Help Output: offset header-col
2746 Column in which "group headers" are printed. A group header is a
2747 descriptive text preceding an option group. For example, in the
2750 Main operation mode:
2752 -A, --catenate, --concatenate append tar files to
2754 -c, --create create a new archive
2755 `Main operation mode:' is the group header.
2757 The default value is 1.
2759 -- Help Output: offset usage-indent
2760 Indentation of wrapped usage lines. Affects `--usage' output.
2763 -- Help Output: offset rmargin
2764 Right margin of the text output. Used for wrapping.
2767 File: tar.info, Node: Fixing Snapshot Files, Next: Tar Internals, Prev: Configuring Help Summary, Up: Top
2769 Appendix C Fixing Snapshot Files
2770 ********************************
2772 Sometimes device numbers can change after upgrading your kernel version
2773 or recofiguring the harvare. Reportedly this is the case with some
2774 newer Linux kernels, when using LVM. In majority of cases this change
2775 is unnoticed by the users. However, it influences `tar' incremental
2776 backups: the device number is stored in tar snapshot files (*note
2777 Snapshot Files::) and is used to determine whether the file has changed
2778 since the last backup. If the device numbers change for some reason,
2779 the next backup you run will be a full backup.
2781 To minimize the impact in these cases, GNU `tar' comes with the
2782 `tar-snapshot-edit' utility for inspecting and updating device numbers
2783 in snapshot files. The utility, written by Dustin J. Mitchell, is
2784 available from GNU `tar' home page
2785 (http://www.gnu.org/software/tar/utils/tar-snapshot-edit.html).
2787 To obtain the device numbers used in the snapshot file, run
2789 $ tar-snapshot-edit SNAPFILE
2791 where SNAPFILE is the name of the snapshot file (you can supply as many
2792 files as you wish in a single command line ).
2794 To update all occurrences of the given device number in the file, use
2795 `-r' option. It takes a single argument of the form `OLDDEV-NEWDEV',
2796 where OLDDEV is the device number used in the snapshot file, and NEWDEV
2797 is the corresponding new device number. Both numbers may be specified
2798 in hex (e.g., `0xfe01'), decimal (e.g., `65025'), or as a major:minor
2799 number pair (e.g., `254:1'). To change several device numbers at once,
2800 specify them in a single comma-separated list, as in `-r
2801 0x3060-0x4500,0x307-0x4600'.
2803 Before updating the snapshot file, it is a good idea to create a
2804 backup copy of it. This is accomplished by `-b' option. The name of
2805 the backup file is obtained by appending `~' to the original file name.
2808 $ tar-snapshot-edit /var/backup/snap.a
2810 /tmp/snap: Device 0x0306 occurs 634 times.
2811 $ tar-snapshot-edit -b -r 0x0306-0x4500 /var/backup/snap.a
2815 File: tar.info, Node: Tar Internals, Next: Genfile, Prev: Fixing Snapshot Files, Up: Top
2817 Appendix D Tar Internals
2818 ************************
2822 * Standard:: Basic Tar Format
2823 * Extensions:: GNU Extensions to the Archive Format
2824 * Sparse Formats:: Storing Sparse Files
2829 File: tar.info, Node: Standard, Next: Extensions, Up: Tar Internals
2834 _(This message will disappear, once this node revised.)_
2836 While an archive may contain many files, the archive itself is a single
2837 ordinary file. Like any other file, an archive file can be written to
2838 a storage device such as a tape or disk, sent through a pipe or over a
2839 network, saved on the active file system, or even stored in another
2840 archive. An archive file is not easy to read or manipulate without
2841 using the `tar' utility or Tar mode in GNU Emacs.
2843 Physically, an archive consists of a series of file entries
2844 terminated by an end-of-archive entry, which consists of two 512 blocks
2845 of zero bytes. A file entry usually describes one of the files in the
2846 archive (an "archive member"), and consists of a file header and the
2847 contents of the file. File headers contain file names and statistics,
2848 checksum information which `tar' uses to detect file corruption, and
2849 information about file types.
2851 Archives are permitted to have more than one member with the same
2852 member name. One way this situation can occur is if more than one
2853 version of a file has been stored in the archive. For information
2854 about adding new versions of a file to an archive, see *note update::.
2856 In addition to entries describing archive members, an archive may
2857 contain entries which `tar' itself uses to store information. *Note
2858 label::, for an example of such an archive entry.
2860 A `tar' archive file contains a series of blocks. Each block
2861 contains `BLOCKSIZE' bytes. Although this format may be thought of as
2862 being on magnetic tape, other media are often used.
2864 Each file archived is represented by a header block which describes
2865 the file, followed by zero or more blocks which give the contents of
2866 the file. At the end of the archive file there are two 512-byte blocks
2867 filled with binary zeros as an end-of-file marker. A reasonable system
2868 should write such end-of-file marker at the end of an archive, but must
2869 not assume that such a block exists when reading an archive. In
2870 particular GNU `tar' always issues a warning if it does not encounter
2873 The blocks may be "blocked" for physical I/O operations. Each
2874 record of N blocks (where N is set by the `--blocking-factor=512-SIZE'
2875 (`-b 512-SIZE') option to `tar') is written with a single `write ()'
2876 operation. On magnetic tapes, the result of such a write is a single
2877 record. When writing an archive, the last record of blocks should be
2878 written at the full size, with blocks after the zero block containing
2879 all zeros. When reading an archive, a reasonable system should
2880 properly handle an archive whose last record is shorter than the rest,
2881 or which contains garbage records after a zero block.
2883 The header block is defined in C as follows. In the GNU `tar'
2884 distribution, this is part of file `src/tar.h':
2887 /* tar Header Block, from POSIX 1003.1-1990. */
2893 char name[100]; /* 0 */
2894 char mode[8]; /* 100 */
2895 char uid[8]; /* 108 */
2896 char gid[8]; /* 116 */
2897 char size[12]; /* 124 */
2898 char mtime[12]; /* 136 */
2899 char chksum[8]; /* 148 */
2900 char typeflag; /* 156 */
2901 char linkname[100]; /* 157 */
2902 char magic[6]; /* 257 */
2903 char version[2]; /* 263 */
2904 char uname[32]; /* 265 */
2905 char gname[32]; /* 297 */
2906 char devmajor[8]; /* 329 */
2907 char devminor[8]; /* 337 */
2908 char prefix[155]; /* 345 */
2912 #define TMAGIC "ustar" /* ustar and a null */
2914 #define TVERSION "00" /* 00 and no null */
2917 /* Values used in typeflag field. */
2918 #define REGTYPE '0' /* regular file */
2919 #define AREGTYPE '\0' /* regular file */
2920 #define LNKTYPE '1' /* link */
2921 #define SYMTYPE '2' /* reserved */
2922 #define CHRTYPE '3' /* character special */
2923 #define BLKTYPE '4' /* block special */
2924 #define DIRTYPE '5' /* directory */
2925 #define FIFOTYPE '6' /* FIFO special */
2926 #define CONTTYPE '7' /* reserved */
2928 #define XHDTYPE 'x' /* Extended header referring to the
2929 next file in the archive */
2930 #define XGLTYPE 'g' /* Global extended header */
2932 /* Bits used in the mode field, values in octal. */
2933 #define TSUID 04000 /* set UID on execution */
2934 #define TSGID 02000 /* set GID on execution */
2935 #define TSVTX 01000 /* reserved */
2936 /* file permissions */
2937 #define TUREAD 00400 /* read by owner */
2938 #define TUWRITE 00200 /* write by owner */
2939 #define TUEXEC 00100 /* execute/search by owner */
2940 #define TGREAD 00040 /* read by group */
2941 #define TGWRITE 00020 /* write by group */
2942 #define TGEXEC 00010 /* execute/search by group */
2943 #define TOREAD 00004 /* read by other */
2944 #define TOWRITE 00002 /* write by other */
2945 #define TOEXEC 00001 /* execute/search by other */
2947 /* tar Header Block, GNU extensions. */
2949 /* In GNU tar, SYMTYPE is for to symbolic links, and CONTTYPE is for
2950 contiguous files, so maybe disobeying the `reserved' comment in POSIX
2951 header description. I suspect these were meant to be used this way, and
2952 should not have really been `reserved' in the published standards. */
2954 /* *BEWARE* *BEWARE* *BEWARE* that the following information is still
2955 boiling, and may change. Even if the OLDGNU format description should be
2956 accurate, the so-called GNU format is not yet fully decided. It is
2957 surely meant to use only extensions allowed by POSIX, but the sketch
2958 below repeats some ugliness from the OLDGNU format, which should rather
2959 go away. Sparse files should be saved in such a way that they do *not*
2960 require two passes at archive creation time. Huge files get some POSIX
2961 fields to overflow, alternate solutions have to be sought for this. */
2963 /* Descriptor for a single file hole. */
2967 char offset[12]; /* 0 */
2968 char numbytes[12]; /* 12 */
2972 /* Sparse files are not supported in POSIX ustar format. For sparse files
2973 with a POSIX header, a GNU extra header is provided which holds overall
2974 sparse information and a few sparse descriptors. When an old GNU header
2975 replaces both the POSIX header and the GNU extra header, it holds some
2976 sparse descriptors too. Whether POSIX or not, if more sparse descriptors
2977 are still needed, they are put into as many successive sparse headers as
2978 necessary. The following constants tell how many sparse descriptors fit
2979 in each kind of header able to hold them. */
2981 #define SPARSES_IN_EXTRA_HEADER 16
2982 #define SPARSES_IN_OLDGNU_HEADER 4
2983 #define SPARSES_IN_SPARSE_HEADER 21
2985 /* Extension header for sparse files, used immediately after the GNU extra
2986 header, and used only if all sparse information cannot fit into that
2987 extra header. There might even be many such extension headers, one after
2988 the other, until all sparse information has been recorded. */
2990 struct sparse_header
2992 struct sparse sp[SPARSES_IN_SPARSE_HEADER];
2994 char isextended; /* 504 */
2998 /* The old GNU format header conflicts with POSIX format in such a way that
2999 POSIX archives may fool old GNU tar's, and POSIX tar's might well be
3000 fooled by old GNU tar archives. An old GNU format header uses the space
3001 used by the prefix field in a POSIX header, and cumulates information
3002 normally found in a GNU extra header. With an old GNU tar header, we
3003 never see any POSIX header nor GNU extra header. Supplementary sparse
3004 headers are allowed, however. */
3006 struct oldgnu_header
3008 char unused_pad1[345]; /* 0 */
3009 char atime[12]; /* 345 Incr. archive: atime of the file */
3010 char ctime[12]; /* 357 Incr. archive: ctime of the file */
3011 char offset[12]; /* 369 Multivolume archive: the offset of
3012 the start of this volume */
3013 char longnames[4]; /* 381 Not used */
3014 char unused_pad2; /* 385 */
3015 struct sparse sp[SPARSES_IN_OLDGNU_HEADER];
3017 char isextended; /* 482 Sparse file: Extension sparse header
3019 char realsize[12]; /* 483 Sparse file: Real size*/
3023 /* OLDGNU_MAGIC uses both magic and version fields, which are contiguous.
3024 Found in an archive, it indicates an old GNU header format, which will be
3025 hopefully become obsolescent. With OLDGNU_MAGIC, uname and gname are
3026 valid, though the header is not truly POSIX conforming. */
3027 #define OLDGNU_MAGIC "ustar " /* 7 chars and a null */
3029 /* The standards committee allows only capital A through capital Z for
3030 user-defined expansion. Other letters in use include:
3032 'A' Solaris Access Control List
3033 'E' Solaris Extended Attribute File
3034 'I' Inode only, as in 'star'
3035 'N' Obsolete GNU tar, for file names that do not fit into the main header.
3036 'X' POSIX 1003.1-2001 eXtended (VU version) */
3038 /* This is a dir entry that contains the names of files that were in the
3039 dir at the time the dump was made. */
3040 #define GNUTYPE_DUMPDIR 'D'
3042 /* Identifies the *next* file on the tape as having a long linkname. */
3043 #define GNUTYPE_LONGLINK 'K'
3045 /* Identifies the *next* file on the tape as having a long name. */
3046 #define GNUTYPE_LONGNAME 'L'
3048 /* This is the continuation of a file that began on another volume. */
3049 #define GNUTYPE_MULTIVOL 'M'
3051 /* This is for sparse files. */
3052 #define GNUTYPE_SPARSE 'S'
3054 /* This file is a tape/volume header. Ignore it on extraction. */
3055 #define GNUTYPE_VOLHDR 'V'
3057 /* Solaris extended header */
3058 #define SOLARIS_XHDTYPE 'X'
3060 /* Jo"rg Schilling star header */
3064 char name[100]; /* 0 */
3065 char mode[8]; /* 100 */
3066 char uid[8]; /* 108 */
3067 char gid[8]; /* 116 */
3068 char size[12]; /* 124 */
3069 char mtime[12]; /* 136 */
3070 char chksum[8]; /* 148 */
3071 char typeflag; /* 156 */
3072 char linkname[100]; /* 157 */
3073 char magic[6]; /* 257 */
3074 char version[2]; /* 263 */
3075 char uname[32]; /* 265 */
3076 char gname[32]; /* 297 */
3077 char devmajor[8]; /* 329 */
3078 char devminor[8]; /* 337 */
3079 char prefix[131]; /* 345 */
3080 char atime[12]; /* 476 */
3081 char ctime[12]; /* 488 */
3085 #define SPARSES_IN_STAR_HEADER 4
3086 #define SPARSES_IN_STAR_EXT_HEADER 21
3088 struct star_in_header
3090 char fill[345]; /* 0 Everything that is before t_prefix */
3091 char prefix[1]; /* 345 t_name prefix */
3092 char fill2; /* 346 */
3093 char fill3[8]; /* 347 */
3094 char isextended; /* 355 */
3095 struct sparse sp[SPARSES_IN_STAR_HEADER]; /* 356 */
3096 char realsize[12]; /* 452 Actual size of the file */
3097 char offset[12]; /* 464 Offset of multivolume contents */
3098 char atime[12]; /* 476 */
3099 char ctime[12]; /* 488 */
3100 char mfill[8]; /* 500 */
3101 char xmagic[4]; /* 508 "tar" */
3104 struct star_ext_header
3106 struct sparse sp[SPARSES_IN_STAR_EXT_HEADER];
3110 All characters in header blocks are represented by using 8-bit
3111 characters in the local variant of ASCII. Each field within the
3112 structure is contiguous; that is, there is no padding used within the
3113 structure. Each character on the archive medium is stored contiguously.
3115 Bytes representing the contents of files (after the header block of
3116 each file) are not translated in any way and are not constrained to
3117 represent characters in any character set. The `tar' format does not
3118 distinguish text files from binary files, and no translation of file
3119 contents is performed.
3121 The `name', `linkname', `magic', `uname', and `gname' are
3122 null-terminated character strings. All other fields are zero-filled
3123 octal numbers in ASCII. Each numeric field of width W contains W minus
3124 1 digits, and a null.
3126 The `name' field is the file name of the file, with directory names
3127 (if any) preceding the file name, separated by slashes.
3129 The `mode' field provides nine bits specifying file permissions and
3130 three bits to specify the Set UID, Set GID, and Save Text ("sticky")
3131 modes. Values for these bits are defined above. When special
3132 permissions are required to create a file with a given mode, and the
3133 user restoring files from the archive does not hold such permissions,
3134 the mode bit(s) specifying those special permissions are ignored.
3135 Modes which are not supported by the operating system restoring files
3136 from the archive will be ignored. Unsupported modes should be faked up
3137 when creating or updating an archive; e.g., the group permission could
3138 be copied from the _other_ permission.
3140 The `uid' and `gid' fields are the numeric user and group ID of the
3141 file owners, respectively. If the operating system does not support
3142 numeric user or group IDs, these fields should be ignored.
3144 The `size' field is the size of the file in bytes; linked files are
3145 archived with this field specified as zero.
3147 The `mtime' field is the data modification time of the file at the
3148 time it was archived. It is the ASCII representation of the octal
3149 value of the last time the file's contents were modified, represented
3150 as an integer number of seconds since January 1, 1970, 00:00
3151 Coordinated Universal Time.
3153 The `chksum' field is the ASCII representation of the octal value of
3154 the simple sum of all bytes in the header block. Each 8-bit byte in
3155 the header is added to an unsigned integer, initialized to zero, the
3156 precision of which shall be no less than seventeen bits. When
3157 calculating the checksum, the `chksum' field is treated as if it were
3160 The `typeflag' field specifies the type of file archived. If a
3161 particular implementation does not recognize or permit the specified
3162 type, the file will be extracted as if it were a regular file. As this
3163 action occurs, `tar' issues a warning to the standard error.
3165 The `atime' and `ctime' fields are used in making incremental
3166 backups; they store, respectively, the particular file's access and
3167 status change times.
3169 The `offset' is used by the `--multi-volume' (`-M') option, when
3170 making a multi-volume archive. The offset is number of bytes into the
3171 file that we need to restart at to continue the file on the next tape,
3172 i.e., where we store the location that a continued file is continued at.
3174 The following fields were added to deal with sparse files. A file
3175 is "sparse" if it takes in unallocated blocks which end up being
3176 represented as zeros, i.e., no useful data. A test to see if a file is
3177 sparse is to look at the number blocks allocated for it versus the
3178 number of characters in the file; if there are fewer blocks allocated
3179 for the file than would normally be allocated for a file of that size,
3180 then the file is sparse. This is the method `tar' uses to detect a
3181 sparse file, and once such a file is detected, it is treated
3182 differently from non-sparse files.
3184 Sparse files are often `dbm' files, or other database-type files
3185 which have data at some points and emptiness in the greater part of the
3186 file. Such files can appear to be very large when an `ls -l' is done
3187 on them, when in truth, there may be a very small amount of important
3188 data contained in the file. It is thus undesirable to have `tar' think
3189 that it must back up this entire file, as great quantities of room are
3190 wasted on empty blocks, which can lead to running out of room on a tape
3191 far earlier than is necessary. Thus, sparse files are dealt with so
3192 that these empty blocks are not written to the tape. Instead, what is
3193 written to the tape is a description, of sorts, of the sparse file:
3194 where the holes are, how big the holes are, and how much data is found
3195 at the end of the hole. This way, the file takes up potentially far
3196 less room on the tape, and when the file is extracted later on, it will
3197 look exactly the way it looked beforehand. The following is a
3198 description of the fields used to handle a sparse file:
3200 The `sp' is an array of `struct sparse'. Each `struct sparse'
3201 contains two 12-character strings which represent an offset into the
3202 file and a number of bytes to be written at that offset. The offset is
3203 absolute, and not relative to the offset in preceding array element.
3205 The header can hold four of these `struct sparse' at the moment; if
3206 more are needed, they are not stored in the header.
3208 The `isextended' flag is set when an `extended_header' is needed to
3209 deal with a file. Note that this means that this flag can only be set
3210 when dealing with a sparse file, and it is only set in the event that
3211 the description of the file will not fit in the allotted room for
3212 sparse structures in the header. In other words, an extended_header is
3215 The `extended_header' structure is used for sparse files which need
3216 more sparse structures than can fit in the header. The header can fit
3217 4 such structures; if more are needed, the flag `isextended' gets set
3218 and the next block is an `extended_header'.
3220 Each `extended_header' structure contains an array of 21 sparse
3221 structures, along with a similar `isextended' flag that the header had.
3222 There can be an indeterminate number of such `extended_header's to
3223 describe a sparse file.
3227 These flags represent a regular file. In order to be compatible
3228 with older versions of `tar', a `typeflag' value of `AREGTYPE'
3229 should be silently recognized as a regular file. New archives
3230 should be created using `REGTYPE'. Also, for backward
3231 compatibility, `tar' treats a regular file whose name ends with a
3232 slash as a directory.
3235 This flag represents a file linked to another file, of any type,
3236 previously archived. Such files are identified in Unix by each
3237 file having the same device and inode number. The linked-to name
3238 is specified in the `linkname' field with a trailing null.
3241 This represents a symbolic link to another file. The linked-to
3242 name is specified in the `linkname' field with a trailing null.
3246 These represent character special files and block special files
3247 respectively. In this case the `devmajor' and `devminor' fields
3248 will contain the major and minor device numbers respectively.
3249 Operating systems may map the device specifications to their own
3250 local specification, or may ignore the entry.
3253 This flag specifies a directory or sub-directory. The directory
3254 name in the `name' field should end with a slash. On systems where
3255 disk allocation is performed on a directory basis, the `size' field
3256 will contain the maximum number of bytes (which may be rounded to
3257 the nearest disk block allocation unit) which the directory may
3258 hold. A `size' field of zero indicates no such limiting. Systems
3259 which do not support limiting in this manner should ignore the
3263 This specifies a FIFO special file. Note that the archiving of a
3264 FIFO file archives the existence of this file and not its contents.
3267 This specifies a contiguous file, which is the same as a normal
3268 file except that, in operating systems which support it, all its
3269 space is allocated contiguously on the disk. Operating systems
3270 which do not allow contiguous allocation should silently treat this
3271 type as a normal file.
3274 These are reserved for custom implementations. Some of these are
3275 used in the GNU modified format, as described below.
3278 Other values are reserved for specification in future revisions of
3279 the P1003 standard, and should not be used by any `tar' program.
3281 The `magic' field indicates that this archive was output in the
3282 P1003 archive format. If this field contains `TMAGIC', the `uname' and
3283 `gname' fields will contain the ASCII representation of the owner and
3284 group of the file respectively. If found, the user and group IDs are
3285 used rather than the values in the `uid' and `gid' fields.
3287 For references, see ISO/IEC 9945-1:1990 or IEEE Std 1003.1-1990,
3288 pages 169-173 (section 10.1) for `Archive/Interchange File Format'; and
3289 IEEE Std 1003.2-1992, pages 380-388 (section 4.48) and pages 936-940
3290 (section E.4.48) for `pax - Portable archive interchange'.
3293 File: tar.info, Node: Extensions, Next: Sparse Formats, Prev: Standard, Up: Tar Internals
3295 GNU Extensions to the Archive Format
3296 ====================================
3298 _(This message will disappear, once this node revised.)_
3300 The GNU format uses additional file types to describe new types of
3301 files in an archive. These are listed below.
3305 This represents a directory and a list of files created by the
3306 `--incremental' (`-G') option. The `size' field gives the total
3307 size of the associated list of files. Each file name is preceded
3308 by either a `Y' (the file should be in this archive) or an `N'.
3309 (The file is a directory, or is not stored in the archive.) Each
3310 file name is terminated by a null. There is an additional null
3311 after the last file name.
3315 This represents a file continued from another volume of a
3316 multi-volume archive created with the `--multi-volume' (`-M')
3317 option. The original type of the file is not given here. The
3318 `size' field gives the maximum size of this piece of the file
3319 (assuming the volume does not end before the file is written out).
3320 The `offset' field gives the offset from the beginning of the file
3321 where this part of the file begins. Thus `size' plus `offset'
3322 should equal the original size of the file.
3326 This flag indicates that we are dealing with a sparse file. Note
3327 that archiving a sparse file requires special operations to find
3328 holes in the file, which mark the positions of these holes, along
3329 with the number of bytes of data to be found after the hole.
3333 This file type is used to mark the volume header that was given
3334 with the `--label=ARCHIVE-LABEL' (`-V ARCHIVE-LABEL') option when
3335 the archive was created. The `name' field contains the `name'
3336 given after the `--label=ARCHIVE-LABEL' (`-V ARCHIVE-LABEL')
3337 option. The `size' field is zero. Only the first file in each
3338 volume of an archive should have this type.
3341 You may have trouble reading a GNU format archive on a non-GNU
3342 system if the options `--incremental' (`-G'), `--multi-volume' (`-M'),
3343 `--sparse' (`-S'), or `--label=ARCHIVE-LABEL' (`-V ARCHIVE-LABEL') were
3344 used when writing the archive. In general, if `tar' does not use the
3345 GNU-added fields of the header, other versions of `tar' should be able
3346 to read the archive. Otherwise, the `tar' program will give an error,
3347 the most likely one being a checksum error.
3350 File: tar.info, Node: Sparse Formats, Next: Snapshot Files, Prev: Extensions, Up: Tar Internals
3352 Storing Sparse Files
3353 ====================
3355 The notion of sparse file, and the ways of handling it from the point
3356 of view of GNU `tar' user have been described in detail in *note
3357 sparse::. This chapter describes the internal format GNU `tar' uses to
3360 The support for sparse files in GNU `tar' has a long history. The
3361 earliest version featuring this support that I was able to find was
3362 1.09, released in November, 1990. The format introduced back then is
3363 called "old GNU" sparse format and in spite of the fact that its design
3364 contained many flaws, it was the only format GNU `tar' supported until
3365 version 1.14 (May, 2004), which introduced initial support for sparse
3366 archives in PAX archives (*note posix::). This format was not free
3367 from design flows, either and it was subsequently improved in versions
3368 1.15.2 (November, 2005) and 1.15.92 (June, 2006).
3370 In addition to GNU sparse format, GNU `tar' is able to read and
3371 extract sparse files archived by `star'.
3373 The following subsections describe each format in detail.
3378 * PAX 0:: PAX Format, Versions 0.0 and 0.1
3379 * PAX 1:: PAX Format, Version 1.0
3382 File: tar.info, Node: Old GNU Format, Next: PAX 0, Up: Sparse Formats
3384 D.0.1 Old GNU Format
3385 --------------------
3387 The format introduced some time around 1990 (v. 1.09). It was designed
3388 on top of standard `ustar' headers in such an unfortunate way that some
3389 of its fields overwrote fields required by POSIX.
3391 An old GNU sparse header is designated by type `S'
3392 (`GNUTYPE_SPARSE') and has the following layout:
3394 Offset Size Name Data type Contents
3395 ----------------------------------------------------------------------------
3397 345 12 atime Number `atime' of the file.
3398 357 12 ctime Number `ctime' of the file .
3399 369 12 offset Number For multivolume archives:
3400 the offset of the start of
3404 386 96 sp `sparse_header'(4 entries) File map.
3405 482 1 isextended Bool `1' if an extension sparse
3408 483 12 realsize Number Real size of the file.
3410 Each of `sparse_header' object at offset 386 describes a single data
3411 chunk. It has the following structure:
3413 Offset Size Data type Contents
3414 ---------------------------------------------------------------------------
3415 0 12 Number Offset of the beginning of the chunk.
3416 12 12 Number Size of the chunk.
3418 If the member contains more than four chunks, the `isextended' field
3419 of the header has the value `1' and the main header is followed by one
3420 or more "extension headers". Each such header has the following
3423 Offset Size Name Data type Contents
3424 ----------------------------------------------------------------------------
3425 0 21 sp `sparse_header' (21 entires) File map.
3426 504 1 isextended Bool `1' if an extension sparse
3427 header follows, or `0'
3430 A header with `isextended=0' ends the map.
3433 File: tar.info, Node: PAX 0, Next: PAX 1, Prev: Old GNU Format, Up: Sparse Formats
3435 D.0.2 PAX Format, Versions 0.0 and 0.1
3436 --------------------------------------
3438 There are two formats available in this branch. The version `0.0' is
3439 the initial version of sparse format used by `tar' versions
3440 1.14-1.15.1. The sparse file map is kept in extended (`x') PAX header
3444 Real size of the stored file
3446 `GNU.sparse.numblocks'
3447 Number of blocks in the sparse map
3450 Offset of the data block
3452 `GNU.sparse.numbytes'
3453 Size of the data block
3455 The latter two variables repeat for each data block, so the overall
3456 structure is like this:
3458 GNU.sparse.size=SIZE
3459 GNU.sparse.numblocks=NUMBLOCKS
3460 repeat NUMBLOCKS times
3461 GNU.sparse.offset=OFFSET
3462 GNU.sparse.numbytes=NUMBYTES
3465 This format presented the following two problems:
3467 1. Whereas the POSIX specification allows a variable to appear
3468 multiple times in a header, it requires that only the last
3469 occurrence be meaningful. Thus, multiple occurrences of
3470 `GNU.sparse.offset' and `GNU.sparse.numbytes' are conflicting with
3473 2. Attempting to extract such archives using a third-party `tar's
3474 results in extraction of sparse files in _compressed form_. If
3475 the `tar' implementation in question does not support POSIX
3476 format, it will also extract a file containing extension header
3477 attributes. This file can be used to expand the file to its
3478 original state. However, posix-aware `tar's will usually ignore
3479 the unknown variables, which makes restoring the file more
3480 difficult. *Note Extraction of sparse members in v.0.0 format:
3481 extracting sparse v.0.x, for the detailed description of how to
3482 restore such members using non-GNU `tar's.
3484 GNU `tar' 1.15.2 introduced sparse format version `0.1', which
3485 attempted to solve these problems. As its predecessor, this format
3486 stores sparse map in the extended POSIX header. It retains
3487 `GNU.sparse.size' and `GNU.sparse.numblocks' variables, but instead of
3488 `GNU.sparse.offset'/`GNU.sparse.numbytes' pairs it uses a single
3492 Map of non-null data chunks. It is a string consisting of
3493 comma-separated values "OFFSET,SIZE[,OFFSET-1,SIZE-1...]"
3495 To address the 2nd problem, the `name' field in `ustar' is replaced
3496 with a special name, constructed using the following pattern:
3498 %d/GNUSparseFile.%p/%f
3500 The real name of the sparse file is stored in the variable
3501 `GNU.sparse.name'. Thus, those `tar' implementations that are not
3502 aware of GNU extensions will at least extract the files into separate
3503 directories, giving the user a possibility to expand it afterwards.
3504 *Note Extraction of sparse members in v.0.1 format: extracting sparse
3505 v.0.x, for the detailed description of how to restore such members
3506 using non-GNU `tar's.
3508 The resulting `GNU.sparse.map' string can be _very_ long. Although
3509 POSIX does not impose any limit on the length of a `x' header variable,
3510 this possibly can confuse some tars.
3513 File: tar.info, Node: PAX 1, Prev: PAX 0, Up: Sparse Formats
3515 D.0.3 PAX Format, Version 1.0
3516 -----------------------------
3518 The version `1.0' of sparse format was introduced with GNU `tar'
3519 1.15.92. Its main objective was to make the resulting file extractable
3520 with little effort even by non-posix aware `tar' implementations.
3521 Starting from this version, the extended header preceding a sparse
3522 member always contains the following variables that identify the format
3531 The `name' field in `ustar' header contains a special name,
3532 constructed using the following pattern:
3534 %d/GNUSparseFile.%p/%f
3536 The real name of the sparse file is stored in the variable
3537 `GNU.sparse.name'. The real size of the file is stored in the variable
3538 `GNU.sparse.realsize'.
3540 The sparse map itself is stored in the file data block, preceding
3541 the actual file data. It consists of a series of octal numbers of
3542 arbitrary length, delimited by newlines. The map is padded with nulls
3543 to the nearest block boundary.
3545 The first number gives the number of entries in the map. Following
3546 are map entries, each one consisting of two numbers giving the offset
3547 and size of the data block it describes.
3549 The format is designed in such a way that non-posix aware tars and
3550 tars not supporting `GNU.sparse.*' keywords will extract each sparse
3551 file in its condensed form with the file map prepended and will place it
3552 into a separate directory. Then, using a simple program it would be
3553 possible to expand the file to its original form even without GNU `tar'.
3554 *Note Sparse Recovery::, for the detailed information on how to extract
3555 sparse members without GNU `tar'.
3558 File: tar.info, Node: Snapshot Files, Next: Dumpdir, Prev: Sparse Formats, Up: Tar Internals
3560 Format of the Incremental Snapshot Files
3561 ========================================
3563 A "snapshot file" (or "directory file") is created during incremental
3564 backups (*note Incremental Dumps::). It contains the status of the
3565 file system at the time of the dump and is used to determine which
3566 files were modified since the last backup.
3568 GNU `tar' version 1.21 supports three snapshot file formats. The
3569 first format, called "format 0", is the one used by GNU `tar' versions
3570 up to 1.15.1. The second format, called "format 1" is an extended
3571 version of this format, that contains more metadata and allows for
3572 further extensions. It was used by version 1.15.1. Starting from
3573 version 1.16 and up to 1.21, the "format 2" is used.
3575 GNU `tar' is able to read all three formats, but will create
3576 snapshots only in format 2.
3578 This appendix describes all three formats in detail.
3580 0. `Format 0' snapshot file begins with a line containing a decimal
3581 number that represents a UNIX timestamp of the beginning of the
3582 last archivation. This line is followed by directory metadata
3583 descriptions, one per line. Each description has the following
3591 A single plus character (`+'), if this directory is located on
3592 an NFS-mounted partition, or a single space otherwise;
3595 Device number of the directory;
3598 I-node number of the directory;
3601 Name of the directory. Any special characters (white-space,
3602 backslashes, etc.) are quoted.
3604 1. `Format 1' snapshot file begins with a line specifying the
3605 format of the file. This line has the following structure:
3607 `GNU tar-'TAR-VERSION`-'INCR-FORMAT-VERSION
3609 where TAR-VERSION is the version number of GNU `tar'
3610 implementation that created this snapshot, and INCR-FORMAT-VERSION
3611 is the version number of the snapshot format (in this case `1').
3613 Next line contains two decimal numbers, representing the time of
3614 the last backup. First number is the number of seconds, the second
3615 one is the number of nanoseconds, since the beginning of the epoch.
3617 Lines that follow contain directory metadata, one line per
3618 directory. Each line is formatted as follows:
3620 [NFS]MTIME-SEC MTIME-NSEC DEV INODE NAME
3622 where MTIME-SEC and MTIME-NSEC represent last modification time of
3623 this directory with nanosecond precision; NFS, DEV, INODE and NAME
3624 have the same meaning as with `format 0'.
3626 2. A snapshot file begins with a format identifier, as described
3627 for version 1, e.g.:
3631 This line is followed by newline. Rest of file consists of
3632 records, separated by null (ASCII 0) characters. Thus, in contrast
3633 to the previous formats, format 2 snapshot is a binary file.
3635 First two records are decimal numbers, representing the time of
3636 the last backup. First number is the number of seconds, the
3637 second one is the number of nanoseconds, since the beginning of the
3638 epoch. These are followed by arbitrary number of directory
3641 Each "directory record" contains a set of metadata describing a
3642 particular directory. Parts of a directory record are delimited
3643 with ASCII 0 characters. The following table describes each part.
3644 The "Number" type in this table stands for a decimal number in
3647 Field Type Description
3648 ---------------------------------------------------------------------
3649 nfs Character `1' if the directory is located on an
3650 NFS-mounted partition, or `0' otherwise;
3651 mtime-sec Number Modification time, seconds;
3652 mtime-nano Number Modification time, nanoseconds;
3653 dev-no Number Device number;
3654 i-no Number I-node number;
3655 name String Directory name; In contrast to the
3656 previous versions it is not quoted.
3657 contents Dumpdir Contents of the directory; *Note
3658 Dumpdir::, for a description of its
3662 Dumpdirs stored in snapshot files contain only records of types
3667 File: tar.info, Node: Dumpdir, Prev: Snapshot Files, Up: Tar Internals
3672 Incremental archives keep information about contents of each dumped
3673 directory in special data blocks called "dumpdirs".
3675 Dumpdir is a sequence of entries of the following form:
3679 where C is one of the "control codes" described below, FILENAME is the
3680 name of the file C operates upon, and `\0' represents a nul character
3681 (ASCII 0). The white space characters were added for readability, real
3682 dumpdirs do not contain them.
3684 Each dumpdir ends with a single nul character.
3686 The following table describes control codes and their meanings:
3689 FILENAME is contained in the archive.
3692 FILENAME was present in the directory at the time the archive was
3693 made, yet it was not dumped to the archive, because it had not
3694 changed since the last backup.
3697 FILENAME is a directory.
3700 This code requests renaming of the FILENAME to the name specified
3701 with the `T' command, that immediately follows it.
3704 Specify target file name for `R' command (see below).
3707 Specify "temporary directory" name for a rename operation (see
3710 Codes `Y', `N' and `D' require FILENAME argument to be a relative
3711 file name to the directory this dumpdir describes, whereas codes `R',
3712 `T' and `X' require their argument to be an absolute file name.
3714 The three codes `R', `T' and `X' specify a "renaming operation". In
3715 the simplest case it is:
3717 R`source'\0T`dest'\0
3719 which means "rename file `source' to file `dest'".
3721 However, there are cases that require using a "temporary directory".
3722 For example, consider the following scenario:
3724 1. Previous run dumped a directory `foo' which contained the
3725 following three directories:
3731 2. They were renamed _cyclically_, so that:
3737 3. New incremental dump was made.
3739 This case cannot be handled by three successive renames, since
3740 renaming `a' to `b' will destroy the existing directory. To correctly
3741 process it, GNU `tar' needs a temporary directory, so it creates the
3742 following dumpdir (newlines have been added for readability):
3750 The first command, `Xfoo\0', instructs the extractor to create a
3751 temporary directory in the directory `foo'. Second command,
3752 `Rfoo/aT\0', says "rename file `foo/a' to the temporary directory that
3753 has just been created" (empty file name after a command means use
3754 temporary directory). Third and fourth commands work as usual, and,
3755 finally, the last command, `R\0Tfoo/a\0' tells tar to rename the
3756 temporary directory to `foo/a'.
3758 The exact placement of a dumpdir in the archive depends on the
3759 archive format (*note Formats::):
3763 In PAX archives, dumpdir is stored in the extended header of the
3764 corresponding directory, in variable `GNU.dumpdir'.
3766 * GNU and old GNU archives
3768 These formats implement special header type `D', which is similar
3769 to ustar header `5' (directory), except that it precedes a data
3770 block containing the dumpdir.
3773 File: tar.info, Node: Genfile, Next: Free Software Needs Free Documentation, Prev: Tar Internals, Up: Top
3778 This appendix describes `genfile', an auxiliary program used in the GNU
3779 tar testsuite. If you are not interested in developing GNU tar, skip
3782 Initially, `genfile' was used to generate data files for the
3783 testsuite, hence its name. However, new operation modes were being
3784 implemented as the testsuite grew more sophisticated, and now `genfile'
3785 is a multi-purpose instrument.
3787 There are three basic operation modes:
3790 This is the default mode. In this mode, `genfile' generates data
3794 In this mode `genfile' displays status of specified files.
3796 Synchronous Execution.
3797 In this mode `genfile' executes the given program with
3798 `--checkpoint' option and executes a set of actions when specified
3799 checkpoints are reached.
3803 * Generate Mode:: File Generation Mode.
3804 * Status Mode:: File Status Mode.
3805 * Exec Mode:: Synchronous Execution mode.
3808 File: tar.info, Node: Generate Mode, Next: Status Mode, Up: Genfile
3813 In this mode `genfile' creates a data file for the test suite. The size
3814 of the file is given with the `--length' (`-l') option. By default the
3815 file contents is written to the standard output, this can be changed
3816 using `--file' (`-f') command line option. Thus, the following two
3817 commands are equivalent:
3819 genfile --length 100 > outfile
3820 genfile --length 100 --file outfile
3822 If `--length' is not given, `genfile' will generate an empty
3825 The command line option `--seek=N' istructs `genfile' to skip the
3826 given number of bytes (N) in the output file before writing to it. It
3827 is similar to the `seek=N' of the `dd' utility.
3829 You can instruct `genfile' to create several files at one go, by
3830 giving it `--files-from' (`-T') option followed by a name of file
3831 containing a list of file names. Using dash (`-') instead of the file
3832 name causes `genfile' to read file list from the standard input. For
3835 # Read file names from file `file.list'
3836 genfile --files-from file.list
3837 # Read file names from standard input
3838 genfile --files-from -
3840 The list file is supposed to contain one file name per line. To use
3841 file lists separated by ASCII NUL character, use `--null' (`-0')
3842 command line option:
3844 genfile --null --files-from file.list
3846 The default data pattern for filling the generated file consists of
3847 first 256 letters of ASCII code, repeated enough times to fill the
3848 entire file. This behavior can be changed with `--pattern' option. This
3849 option takes a mandatory argument, specifying pattern name to use.
3850 Currently two patterns are implemented:
3853 The default pattern as described above.
3856 Fills the file with zeroes.
3858 If no file name was given, the program exits with the code `0'.
3859 Otherwise, it exits with `0' only if it was able to create a file of
3860 the specified length.
3862 Special option `--sparse' (`-s') instructs `genfile' to create a
3863 sparse file. Sparse files consist of "data fragments", separated by
3864 "holes" or blocks of zeros. On many operating systems, actual disk
3865 storage is not allocated for holes, but they are counted in the length
3866 of the file. To create a sparse file, `genfile' should know where to
3867 put data fragments, and what data to use to fill them. So, when
3868 `--sparse' is given the rest of the command line specifies a so-called
3871 The file map consists of any number of "fragment descriptors". Each
3872 descriptor is composed of two values: a number, specifying fragment
3873 offset from the end of the previous fragment or, for the very first
3874 fragment, from the beginning of the file, and "contents string", i.e.,
3875 a string of characters, specifying the pattern to fill the fragment
3876 with. File offset can be suffixed with the following quantifiers:
3880 The number is expressed in kilobytes.
3884 The number is expressed in megabytes.
3888 The number is expressed in gigabytes.
3890 For each letter in contents string `genfile' will generate a "block"
3891 of data, filled with this letter and will write it to the fragment. The
3892 size of block is given by `--block-size' option. It defaults to 512.
3893 Thus, if the string consists of N characters, the resulting file
3894 fragment will contain `N*BLOCK-SIZE' of data.
3896 Last fragment descriptor can have only file offset part. In this
3897 case `genfile' will create a hole at the end of the file up to the
3900 For example, consider the following invocation:
3902 genfile --sparse --file sparsefile 0 ABCD 1M EFGHI 2000K
3904 It will create 3101184-bytes long file of the following structure:
3906 Offset Length Contents
3907 0 4*512=2048 Four 512-byte blocks, filled
3908 with letters `A', `B', `C' and
3910 2048 1046528 Zero bytes
3911 1050624 5*512=2560 Five blocks, filled with letters
3912 `E', `F', `G', `H', `I'.
3913 1053184 2048000 Zero bytes
3915 The exit code of `genfile --status' command is `0' only if created
3916 file is actually sparse.
3919 File: tar.info, Node: Status Mode, Next: Exec Mode, Prev: Generate Mode, Up: Genfile
3924 In status mode, `genfile' prints file system status for each file
3925 specified in the command line. This mode is toggled by `--stat' (`-S')
3926 command line option. An optional argument to this option specifies
3927 output "format": a comma-separated list of `struct stat' fields to be
3928 displayed. This list can contain following identifiers :
3935 Device number in decimal.
3943 File mode in octal. Optional NUMBER specifies octal mask to be
3944 applied to the mode before outputting. For example, `--stat
3945 mode.777' will preserve lower nine bits of it. Notice, that you
3946 can use any punctuation character in place of `.'.
3950 Number of hard links.
3962 File size in decimal.
3966 The size in bytes of each file block.
3970 Number of blocks allocated.
3974 Time of last access.
3978 Time of last modification
3982 Time of last status change
3985 A boolean value indicating whether the file is `sparse'.
3987 Modification times are displayed in UTC as UNIX timestamps, unless
3988 suffixed with `H' (for "human-readable"), as in `ctimeH', in which case
3989 usual `tar tv' output format is used.
3991 The default output format is: `name,dev,ino,mode,
3992 nlink,uid,gid,size,blksize,blocks,atime,mtime,ctime'.
3994 For example, the following command will display file names and
3995 corresponding times of last access for each file in the current working
3998 genfile --stat=name,atime *
4001 File: tar.info, Node: Exec Mode, Prev: Status Mode, Up: Genfile
4006 This mode is designed for testing the behavior of `paxutils' commands
4007 when some of the files change during archiving. It is an experimental
4010 The `Exec Mode' is toggled by `--run' command line option (or its
4011 alias `-r'). The argument to this option gives the command line to be
4012 executed. The actual command line is constructed by inserting
4013 `--checkpoint' option between the command name and its first argument
4014 (if any). Due to this, the argument to `--run' may not use traditional
4015 `tar' option syntax, i.e., the following is wrong:
4018 genfile --run 'tar cf foo bar'
4020 Use the following syntax instead:
4022 genfile --run 'tar -cf foo bar'
4024 The rest of command line after `--run' or its equivalent specifies
4025 checkpoint values and actions to be executed upon reaching them.
4026 Checkpoint values are introduced with `--checkpoint' command line
4027 option. Argument to this option is the number of checkpoint in decimal.
4029 Any number of "actions" may be specified after a checkpoint.
4030 Available actions are
4034 Truncate FILE to the size specified by previous `--length' option
4035 (or 0, if it is not given).
4038 Append data to FILE. The size of data and its pattern are given by
4039 previous `--length' and `pattern' options.
4042 Update the access and modification times of FILE. These timestamps
4043 are changed to the current time, unless `--date' option was given,
4044 in which case they are changed to the specified time. Argument to
4045 `--date' option is a date specification in an almost arbitrary
4046 format (*note Date input formats::).
4049 Execute given shell command.
4052 Option `--verbose' instructs `genfile' to print on standard output
4053 notifications about checkpoints being executed and to verbosely
4054 describe exit status of the command.
4056 While the command is being executed its standard output remains
4057 connected to descriptor 1. All messages it prints to file descriptor 2,
4058 except checkpoint notifications, are forwarded to standard error.
4060 `Genfile' exits with the exit status of the executed command.
4063 File: tar.info, Node: Free Software Needs Free Documentation, Next: Copying This Manual, Prev: Genfile, Up: Top
4065 Appendix F Free Software Needs Free Documentation
4066 *************************************************
4068 The biggest deficiency in the free software community today is not in
4069 the software--it is the lack of good free documentation that we can
4070 include with the free software. Many of our most important programs do
4071 not come with free reference manuals and free introductory texts.
4072 Documentation is an essential part of any software package; when an
4073 important free software package does not come with a free manual and a
4074 free tutorial, that is a major gap. We have many such gaps today.
4076 Consider Perl, for instance. The tutorial manuals that people
4077 normally use are non-free. How did this come about? Because the
4078 authors of those manuals published them with restrictive terms--no
4079 copying, no modification, source files not available--which exclude
4080 them from the free software world.
4082 That wasn't the first time this sort of thing happened, and it was
4083 far from the last. Many times we have heard a GNU user eagerly
4084 describe a manual that he is writing, his intended contribution to the
4085 community, only to learn that he had ruined everything by signing a
4086 publication contract to make it non-free.
4088 Free documentation, like free software, is a matter of freedom, not
4089 price. The problem with the non-free manual is not that publishers
4090 charge a price for printed copies--that in itself is fine. (The Free
4091 Software Foundation sells printed copies of manuals, too.) The problem
4092 is the restrictions on the use of the manual. Free manuals are
4093 available in source code form, and give you permission to copy and
4094 modify. Non-free manuals do not allow this.
4096 The criteria of freedom for a free manual are roughly the same as for
4097 free software. Redistribution (including the normal kinds of
4098 commercial redistribution) must be permitted, so that the manual can
4099 accompany every copy of the program, both on-line and on paper.
4101 Permission for modification of the technical content is crucial too.
4102 When people modify the software, adding or changing features, if they
4103 are conscientious they will change the manual too--so they can provide
4104 accurate and clear documentation for the modified program. A manual
4105 that leaves you no choice but to write a new manual to document a
4106 changed version of the program is not really available to our community.
4108 Some kinds of limits on the way modification is handled are
4109 acceptable. For example, requirements to preserve the original
4110 author's copyright notice, the distribution terms, or the list of
4111 authors, are ok. It is also no problem to require modified versions to
4112 include notice that they were modified. Even entire sections that may
4113 not be deleted or changed are acceptable, as long as they deal with
4114 nontechnical topics (like this one). These kinds of restrictions are
4115 acceptable because they don't obstruct the community's normal use of
4118 However, it must be possible to modify all the _technical_ content
4119 of the manual, and then distribute the result in all the usual media,
4120 through all the usual channels. Otherwise, the restrictions obstruct
4121 the use of the manual, it is not free, and we need another manual to
4124 Please spread the word about this issue. Our community continues to
4125 lose manuals to proprietary publishing. If we spread the word that
4126 free software needs free reference manuals and free tutorials, perhaps
4127 the next person who wants to contribute by writing documentation will
4128 realize, before it is too late, that only free manuals contribute to
4129 the free software community.
4131 If you are writing documentation, please insist on publishing it
4132 under the GNU Free Documentation License or another free documentation
4133 license. Remember that this decision requires your approval--you don't
4134 have to let the publisher decide. Some commercial publishers will use
4135 a free license if you insist, but they will not propose the option; it
4136 is up to you to raise the issue and say firmly that this is what you
4137 want. If the publisher you are dealing with refuses, please try other
4138 publishers. If you're not sure whether a proposed license is free,
4139 write to <licensing@gnu.org>.
4141 You can encourage commercial publishers to sell more free, copylefted
4142 manuals and tutorials by buying them, and particularly by buying copies
4143 from the publishers that paid for their writing or for major
4144 improvements. Meanwhile, try to avoid buying non-free documentation at
4145 all. Check the distribution terms of a manual before you buy it, and
4146 insist that whoever seeks your business must respect your freedom.
4147 Check the history of the book, and try reward the publishers that have
4148 paid or pay the authors to work on it.
4150 The Free Software Foundation maintains a list of free documentation
4151 published by other publishers, at
4152 `http://www.fsf.org/doc/other-free-books.html'.
4155 File: tar.info, Node: Copying This Manual, Next: Index of Command Line Options, Prev: Free Software Needs Free Documentation, Up: Top
4157 Appendix G Copying This Manual
4158 ******************************
4162 * GNU Free Documentation License:: License for copying this manual
4165 File: tar.info, Node: GNU Free Documentation License, Up: Copying This Manual
4167 G.1 GNU Free Documentation License
4168 ==================================
4170 Version 1.2, November 2002
4172 Copyright (C) 2000,2001,2002 Free Software Foundation, Inc.
4173 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA
4175 Everyone is permitted to copy and distribute verbatim copies
4176 of this license document, but changing it is not allowed.
4180 The purpose of this License is to make a manual, textbook, or other
4181 functional and useful document "free" in the sense of freedom: to
4182 assure everyone the effective freedom to copy and redistribute it,
4183 with or without modifying it, either commercially or
4184 noncommercially. Secondarily, this License preserves for the
4185 author and publisher a way to get credit for their work, while not
4186 being considered responsible for modifications made by others.
4188 This License is a kind of "copyleft", which means that derivative
4189 works of the document must themselves be free in the same sense.
4190 It complements the GNU General Public License, which is a copyleft
4191 license designed for free software.
4193 We have designed this License in order to use it for manuals for
4194 free software, because free software needs free documentation: a
4195 free program should come with manuals providing the same freedoms
4196 that the software does. But this License is not limited to
4197 software manuals; it can be used for any textual work, regardless
4198 of subject matter or whether it is published as a printed book.
4199 We recommend this License principally for works whose purpose is
4200 instruction or reference.
4202 1. APPLICABILITY AND DEFINITIONS
4204 This License applies to any manual or other work, in any medium,
4205 that contains a notice placed by the copyright holder saying it
4206 can be distributed under the terms of this License. Such a notice
4207 grants a world-wide, royalty-free license, unlimited in duration,
4208 to use that work under the conditions stated herein. The
4209 "Document", below, refers to any such manual or work. Any member
4210 of the public is a licensee, and is addressed as "you". You
4211 accept the license if you copy, modify or distribute the work in a
4212 way requiring permission under copyright law.
4214 A "Modified Version" of the Document means any work containing the
4215 Document or a portion of it, either copied verbatim, or with
4216 modifications and/or translated into another language.
4218 A "Secondary Section" is a named appendix or a front-matter section
4219 of the Document that deals exclusively with the relationship of the
4220 publishers or authors of the Document to the Document's overall
4221 subject (or to related matters) and contains nothing that could
4222 fall directly within that overall subject. (Thus, if the Document
4223 is in part a textbook of mathematics, a Secondary Section may not
4224 explain any mathematics.) The relationship could be a matter of
4225 historical connection with the subject or with related matters, or
4226 of legal, commercial, philosophical, ethical or political position
4229 The "Invariant Sections" are certain Secondary Sections whose
4230 titles are designated, as being those of Invariant Sections, in
4231 the notice that says that the Document is released under this
4232 License. If a section does not fit the above definition of
4233 Secondary then it is not allowed to be designated as Invariant.
4234 The Document may contain zero Invariant Sections. If the Document
4235 does not identify any Invariant Sections then there are none.
4237 The "Cover Texts" are certain short passages of text that are
4238 listed, as Front-Cover Texts or Back-Cover Texts, in the notice
4239 that says that the Document is released under this License. A
4240 Front-Cover Text may be at most 5 words, and a Back-Cover Text may
4241 be at most 25 words.
4243 A "Transparent" copy of the Document means a machine-readable copy,
4244 represented in a format whose specification is available to the
4245 general public, that is suitable for revising the document
4246 straightforwardly with generic text editors or (for images
4247 composed of pixels) generic paint programs or (for drawings) some
4248 widely available drawing editor, and that is suitable for input to
4249 text formatters or for automatic translation to a variety of
4250 formats suitable for input to text formatters. A copy made in an
4251 otherwise Transparent file format whose markup, or absence of
4252 markup, has been arranged to thwart or discourage subsequent
4253 modification by readers is not Transparent. An image format is
4254 not Transparent if used for any substantial amount of text. A
4255 copy that is not "Transparent" is called "Opaque".
4257 Examples of suitable formats for Transparent copies include plain
4258 ASCII without markup, Texinfo input format, LaTeX input format,
4259 SGML or XML using a publicly available DTD, and
4260 standard-conforming simple HTML, PostScript or PDF designed for
4261 human modification. Examples of transparent image formats include
4262 PNG, XCF and JPG. Opaque formats include proprietary formats that
4263 can be read and edited only by proprietary word processors, SGML or
4264 XML for which the DTD and/or processing tools are not generally
4265 available, and the machine-generated HTML, PostScript or PDF
4266 produced by some word processors for output purposes only.
4268 The "Title Page" means, for a printed book, the title page itself,
4269 plus such following pages as are needed to hold, legibly, the
4270 material this License requires to appear in the title page. For
4271 works in formats which do not have any title page as such, "Title
4272 Page" means the text near the most prominent appearance of the
4273 work's title, preceding the beginning of the body of the text.
4275 A section "Entitled XYZ" means a named subunit of the Document
4276 whose title either is precisely XYZ or contains XYZ in parentheses
4277 following text that translates XYZ in another language. (Here XYZ
4278 stands for a specific section name mentioned below, such as
4279 "Acknowledgements", "Dedications", "Endorsements", or "History".)
4280 To "Preserve the Title" of such a section when you modify the
4281 Document means that it remains a section "Entitled XYZ" according
4284 The Document may include Warranty Disclaimers next to the notice
4285 which states that this License applies to the Document. These
4286 Warranty Disclaimers are considered to be included by reference in
4287 this License, but only as regards disclaiming warranties: any other
4288 implication that these Warranty Disclaimers may have is void and
4289 has no effect on the meaning of this License.
4293 You may copy and distribute the Document in any medium, either
4294 commercially or noncommercially, provided that this License, the
4295 copyright notices, and the license notice saying this License
4296 applies to the Document are reproduced in all copies, and that you
4297 add no other conditions whatsoever to those of this License. You
4298 may not use technical measures to obstruct or control the reading
4299 or further copying of the copies you make or distribute. However,
4300 you may accept compensation in exchange for copies. If you
4301 distribute a large enough number of copies you must also follow
4302 the conditions in section 3.
4304 You may also lend copies, under the same conditions stated above,
4305 and you may publicly display copies.
4307 3. COPYING IN QUANTITY
4309 If you publish printed copies (or copies in media that commonly
4310 have printed covers) of the Document, numbering more than 100, and
4311 the Document's license notice requires Cover Texts, you must
4312 enclose the copies in covers that carry, clearly and legibly, all
4313 these Cover Texts: Front-Cover Texts on the front cover, and
4314 Back-Cover Texts on the back cover. Both covers must also clearly
4315 and legibly identify you as the publisher of these copies. The
4316 front cover must present the full title with all words of the
4317 title equally prominent and visible. You may add other material
4318 on the covers in addition. Copying with changes limited to the
4319 covers, as long as they preserve the title of the Document and
4320 satisfy these conditions, can be treated as verbatim copying in
4323 If the required texts for either cover are too voluminous to fit
4324 legibly, you should put the first ones listed (as many as fit
4325 reasonably) on the actual cover, and continue the rest onto
4328 If you publish or distribute Opaque copies of the Document
4329 numbering more than 100, you must either include a
4330 machine-readable Transparent copy along with each Opaque copy, or
4331 state in or with each Opaque copy a computer-network location from
4332 which the general network-using public has access to download
4333 using public-standard network protocols a complete Transparent
4334 copy of the Document, free of added material. If you use the
4335 latter option, you must take reasonably prudent steps, when you
4336 begin distribution of Opaque copies in quantity, to ensure that
4337 this Transparent copy will remain thus accessible at the stated
4338 location until at least one year after the last time you
4339 distribute an Opaque copy (directly or through your agents or
4340 retailers) of that edition to the public.
4342 It is requested, but not required, that you contact the authors of
4343 the Document well before redistributing any large number of
4344 copies, to give them a chance to provide you with an updated
4345 version of the Document.
4349 You may copy and distribute a Modified Version of the Document
4350 under the conditions of sections 2 and 3 above, provided that you
4351 release the Modified Version under precisely this License, with
4352 the Modified Version filling the role of the Document, thus
4353 licensing distribution and modification of the Modified Version to
4354 whoever possesses a copy of it. In addition, you must do these
4355 things in the Modified Version:
4357 A. Use in the Title Page (and on the covers, if any) a title
4358 distinct from that of the Document, and from those of
4359 previous versions (which should, if there were any, be listed
4360 in the History section of the Document). You may use the
4361 same title as a previous version if the original publisher of
4362 that version gives permission.
4364 B. List on the Title Page, as authors, one or more persons or
4365 entities responsible for authorship of the modifications in
4366 the Modified Version, together with at least five of the
4367 principal authors of the Document (all of its principal
4368 authors, if it has fewer than five), unless they release you
4369 from this requirement.
4371 C. State on the Title page the name of the publisher of the
4372 Modified Version, as the publisher.
4374 D. Preserve all the copyright notices of the Document.
4376 E. Add an appropriate copyright notice for your modifications
4377 adjacent to the other copyright notices.
4379 F. Include, immediately after the copyright notices, a license
4380 notice giving the public permission to use the Modified
4381 Version under the terms of this License, in the form shown in
4384 G. Preserve in that license notice the full lists of Invariant
4385 Sections and required Cover Texts given in the Document's
4388 H. Include an unaltered copy of this License.
4390 I. Preserve the section Entitled "History", Preserve its Title,
4391 and add to it an item stating at least the title, year, new
4392 authors, and publisher of the Modified Version as given on
4393 the Title Page. If there is no section Entitled "History" in
4394 the Document, create one stating the title, year, authors,
4395 and publisher of the Document as given on its Title Page,
4396 then add an item describing the Modified Version as stated in
4397 the previous sentence.
4399 J. Preserve the network location, if any, given in the Document
4400 for public access to a Transparent copy of the Document, and
4401 likewise the network locations given in the Document for
4402 previous versions it was based on. These may be placed in
4403 the "History" section. You may omit a network location for a
4404 work that was published at least four years before the
4405 Document itself, or if the original publisher of the version
4406 it refers to gives permission.
4408 K. For any section Entitled "Acknowledgements" or "Dedications",
4409 Preserve the Title of the section, and preserve in the
4410 section all the substance and tone of each of the contributor
4411 acknowledgements and/or dedications given therein.
4413 L. Preserve all the Invariant Sections of the Document,
4414 unaltered in their text and in their titles. Section numbers
4415 or the equivalent are not considered part of the section
4418 M. Delete any section Entitled "Endorsements". Such a section
4419 may not be included in the Modified Version.
4421 N. Do not retitle any existing section to be Entitled
4422 "Endorsements" or to conflict in title with any Invariant
4425 O. Preserve any Warranty Disclaimers.
4427 If the Modified Version includes new front-matter sections or
4428 appendices that qualify as Secondary Sections and contain no
4429 material copied from the Document, you may at your option
4430 designate some or all of these sections as invariant. To do this,
4431 add their titles to the list of Invariant Sections in the Modified
4432 Version's license notice. These titles must be distinct from any
4433 other section titles.
4435 You may add a section Entitled "Endorsements", provided it contains
4436 nothing but endorsements of your Modified Version by various
4437 parties--for example, statements of peer review or that the text
4438 has been approved by an organization as the authoritative
4439 definition of a standard.
4441 You may add a passage of up to five words as a Front-Cover Text,
4442 and a passage of up to 25 words as a Back-Cover Text, to the end
4443 of the list of Cover Texts in the Modified Version. Only one
4444 passage of Front-Cover Text and one of Back-Cover Text may be
4445 added by (or through arrangements made by) any one entity. If the
4446 Document already includes a cover text for the same cover,
4447 previously added by you or by arrangement made by the same entity
4448 you are acting on behalf of, you may not add another; but you may
4449 replace the old one, on explicit permission from the previous
4450 publisher that added the old one.
4452 The author(s) and publisher(s) of the Document do not by this
4453 License give permission to use their names for publicity for or to
4454 assert or imply endorsement of any Modified Version.
4456 5. COMBINING DOCUMENTS
4458 You may combine the Document with other documents released under
4459 this License, under the terms defined in section 4 above for
4460 modified versions, provided that you include in the combination
4461 all of the Invariant Sections of all of the original documents,
4462 unmodified, and list them all as Invariant Sections of your
4463 combined work in its license notice, and that you preserve all
4464 their Warranty Disclaimers.
4466 The combined work need only contain one copy of this License, and
4467 multiple identical Invariant Sections may be replaced with a single
4468 copy. If there are multiple Invariant Sections with the same name
4469 but different contents, make the title of each such section unique
4470 by adding at the end of it, in parentheses, the name of the
4471 original author or publisher of that section if known, or else a
4472 unique number. Make the same adjustment to the section titles in
4473 the list of Invariant Sections in the license notice of the
4476 In the combination, you must combine any sections Entitled
4477 "History" in the various original documents, forming one section
4478 Entitled "History"; likewise combine any sections Entitled
4479 "Acknowledgements", and any sections Entitled "Dedications". You
4480 must delete all sections Entitled "Endorsements."
4482 6. COLLECTIONS OF DOCUMENTS
4484 You may make a collection consisting of the Document and other
4485 documents released under this License, and replace the individual
4486 copies of this License in the various documents with a single copy
4487 that is included in the collection, provided that you follow the
4488 rules of this License for verbatim copying of each of the
4489 documents in all other respects.
4491 You may extract a single document from such a collection, and
4492 distribute it individually under this License, provided you insert
4493 a copy of this License into the extracted document, and follow
4494 this License in all other respects regarding verbatim copying of
4497 7. AGGREGATION WITH INDEPENDENT WORKS
4499 A compilation of the Document or its derivatives with other
4500 separate and independent documents or works, in or on a volume of
4501 a storage or distribution medium, is called an "aggregate" if the
4502 copyright resulting from the compilation is not used to limit the
4503 legal rights of the compilation's users beyond what the individual
4504 works permit. When the Document is included in an aggregate, this
4505 License does not apply to the other works in the aggregate which
4506 are not themselves derivative works of the Document.
4508 If the Cover Text requirement of section 3 is applicable to these
4509 copies of the Document, then if the Document is less than one half
4510 of the entire aggregate, the Document's Cover Texts may be placed
4511 on covers that bracket the Document within the aggregate, or the
4512 electronic equivalent of covers if the Document is in electronic
4513 form. Otherwise they must appear on printed covers that bracket
4514 the whole aggregate.
4518 Translation is considered a kind of modification, so you may
4519 distribute translations of the Document under the terms of section
4520 4. Replacing Invariant Sections with translations requires special
4521 permission from their copyright holders, but you may include
4522 translations of some or all Invariant Sections in addition to the
4523 original versions of these Invariant Sections. You may include a
4524 translation of this License, and all the license notices in the
4525 Document, and any Warranty Disclaimers, provided that you also
4526 include the original English version of this License and the
4527 original versions of those notices and disclaimers. In case of a
4528 disagreement between the translation and the original version of
4529 this License or a notice or disclaimer, the original version will
4532 If a section in the Document is Entitled "Acknowledgements",
4533 "Dedications", or "History", the requirement (section 4) to
4534 Preserve its Title (section 1) will typically require changing the
4539 You may not copy, modify, sublicense, or distribute the Document
4540 except as expressly provided for under this License. Any other
4541 attempt to copy, modify, sublicense or distribute the Document is
4542 void, and will automatically terminate your rights under this
4543 License. However, parties who have received copies, or rights,
4544 from you under this License will not have their licenses
4545 terminated so long as such parties remain in full compliance.
4547 10. FUTURE REVISIONS OF THIS LICENSE
4549 The Free Software Foundation may publish new, revised versions of
4550 the GNU Free Documentation License from time to time. Such new
4551 versions will be similar in spirit to the present version, but may
4552 differ in detail to address new problems or concerns. See
4553 `http://www.gnu.org/copyleft/'.
4555 Each version of the License is given a distinguishing version
4556 number. If the Document specifies that a particular numbered
4557 version of this License "or any later version" applies to it, you
4558 have the option of following the terms and conditions either of
4559 that specified version or of any later version that has been
4560 published (not as a draft) by the Free Software Foundation. If
4561 the Document does not specify a version number of this License,
4562 you may choose any version ever published (not as a draft) by the
4563 Free Software Foundation.
4565 G.1.1 ADDENDUM: How to use this License for your documents
4566 ----------------------------------------------------------
4568 To use this License in a document you have written, include a copy of
4569 the License in the document and put the following copyright and license
4570 notices just after the title page:
4572 Copyright (C) YEAR YOUR NAME.
4573 Permission is granted to copy, distribute and/or modify this document
4574 under the terms of the GNU Free Documentation License, Version 1.2
4575 or any later version published by the Free Software Foundation;
4576 with no Invariant Sections, no Front-Cover Texts, and no Back-Cover
4577 Texts. A copy of the license is included in the section entitled ``GNU
4578 Free Documentation License''.
4580 If you have Invariant Sections, Front-Cover Texts and Back-Cover
4581 Texts, replace the "with...Texts." line with this:
4583 with the Invariant Sections being LIST THEIR TITLES, with
4584 the Front-Cover Texts being LIST, and with the Back-Cover Texts
4587 If you have Invariant Sections without Cover Texts, or some other
4588 combination of the three, merge those two alternatives to suit the
4591 If your document contains nontrivial examples of program code, we
4592 recommend releasing these examples in parallel under your choice of
4593 free software license, such as the GNU General Public License, to
4594 permit their use in free software.
4597 File: tar.info, Node: Index of Command Line Options, Next: Index, Prev: Copying This Manual, Up: Top
4599 Appendix H Index of Command Line Options
4600 ****************************************
4602 This appendix contains an index of all GNU `tar' long command line
4603 options. The options are listed without the preceding double-dash. For
4604 a cross-reference of short command line options, *note Short Option
4610 * absolute-names: absolute. (line 8)
4611 * absolute-names, summary: Option Summary. (line 6)
4612 * add-file: files. (line 84)
4613 * after-date: after. (line 26)
4614 * after-date, summary: Option Summary. (line 12)
4615 * anchored: controlling pattern-matching.
4617 * anchored, summary: Option Summary. (line 15)
4618 * append: append. (line 8)
4619 * append, summary: Operation Summary. (line 6)
4620 * atime-preserve: Attributes. (line 14)
4621 * atime-preserve, summary: Option Summary. (line 19)
4622 * auto-compress: gzip. (line 76)
4623 * auto-compress, summary: Option Summary. (line 65)
4624 * backup: backup. (line 41)
4625 * backup, summary: Option Summary. (line 71)
4626 * block-number: verbose. (line 115)
4627 * block-number, summary: Option Summary. (line 76)
4628 * blocking-factor: Blocking Factor. (line 8)
4629 * blocking-factor, summary: Option Summary. (line 82)
4630 * bzip2: gzip. (line 130)
4631 * bzip2, summary: Option Summary. (line 87)
4632 * catenate: concatenate. (line 6)
4633 * catenate, summary: Operation Summary. (line 10)
4634 * check-device, described: Incremental Dumps. (line 99)
4635 * check-device, summary: Option Summary. (line 92)
4636 * check-links, described: hard links. (line 33)
4637 * check-links, summary: Option Summary. (line 143)
4638 * checkpoint: checkpoints. (line 6)
4639 * checkpoint, defined: checkpoints. (line 13)
4640 * checkpoint, summary: Option Summary. (line 97)
4641 * checkpoint-action: checkpoints. (line 6)
4642 * checkpoint-action, defined: checkpoints. (line 22)
4643 * checkpoint-action, summary: Option Summary. (line 105)
4644 * compare: compare. (line 8)
4645 * compare, summary: Operation Summary. (line 14)
4646 * compress: gzip. (line 141)
4647 * compress, summary: Option Summary. (line 152)
4648 * concatenate: concatenate. (line 6)
4649 * concatenate, summary: Operation Summary. (line 20)
4650 * confirmation, summary: Option Summary. (line 159)
4651 * create, additional options: create options. (line 6)
4652 * create, complementary notes: Basic tar. (line 11)
4653 * create, introduced: Creating the archive.
4655 * create, summary: Operation Summary. (line 25)
4656 * create, using with --verbose: create verbose. (line 6)
4657 * create, using with --verify: verify. (line 24)
4658 * delay-directory-restore: Directory Modification Times and Permissions.
4660 * delay-directory-restore, summary: Option Summary. (line 162)
4661 * delete: delete. (line 8)
4662 * delete, summary: Operation Summary. (line 29)
4663 * dereference: dereference. (line 6)
4664 * dereference, summary: Option Summary. (line 167)
4665 * diff, summary: Operation Summary. (line 33)
4666 * directory: directory. (line 11)
4667 * directory, summary: Option Summary. (line 173)
4668 * directory, using in --files-from argument: files. (line 60)
4669 * exclude: exclude. (line 11)
4670 * exclude, potential problems with: problems with exclude.
4672 * exclude, summary: Option Summary. (line 180)
4673 * exclude-caches: exclude. (line 96)
4674 * exclude-caches, summary: Option Summary. (line 189)
4675 * exclude-caches-all: exclude. (line 104)
4676 * exclude-caches-all, summary: Option Summary. (line 202)
4677 * exclude-caches-under: exclude. (line 100)
4678 * exclude-caches-under, summary: Option Summary. (line 196)
4679 * exclude-from: exclude. (line 22)
4680 * exclude-from, summary: Option Summary. (line 184)
4681 * exclude-tag: exclude. (line 113)
4682 * exclude-tag, summary: Option Summary. (line 206)
4683 * exclude-tag-all: exclude. (line 121)
4684 * exclude-tag-all, summary: Option Summary. (line 214)
4685 * exclude-tag-under: exclude. (line 117)
4686 * exclude-tag-under, summary: Option Summary. (line 210)
4687 * exclude-vcs: exclude. (line 39)
4688 * exclude-vcs, summary: Option Summary. (line 218)
4689 * extract: extract. (line 8)
4690 * extract, additional options: extract options. (line 8)
4691 * extract, complementary notes: Basic tar. (line 48)
4692 * extract, summary: Operation Summary. (line 37)
4693 * extract, using with --listed-incremental: Incremental Dumps.
4695 * file, short description: file. (line 17)
4696 * file, summary: Option Summary. (line 224)
4697 * file, tutorial: file tutorial. (line 6)
4698 * files-from: files. (line 14)
4699 * files-from, summary: Option Summary. (line 230)
4700 * force-local, short description: Device. (line 70)
4701 * force-local, summary: Option Summary. (line 236)
4702 * format, summary: Option Summary. (line 241)
4703 * get, summary: Operation Summary. (line 42)
4704 * group: override. (line 73)
4705 * group, summary: Option Summary. (line 266)
4706 * gunzip, summary: Option Summary. (line 274)
4707 * gzip: gzip. (line 96)
4708 * gzip, summary: Option Summary. (line 274)
4709 * hard-dereference, described: hard links. (line 61)
4710 * hard-dereference, summary: Option Summary. (line 282)
4711 * help: help tutorial. (line 6)
4712 * help, introduction: help. (line 26)
4713 * help, summary: Option Summary. (line 288)
4714 * ignore-case: controlling pattern-matching.
4716 * ignore-case, summary: Option Summary. (line 293)
4717 * ignore-command-error: Writing to an External Program.
4719 * ignore-command-error, summary: Option Summary. (line 297)
4720 * ignore-failed-read: Ignore Failed Read. (line 7)
4721 * ignore-failed-read, summary: Option Summary. (line 301)
4722 * ignore-zeros: Ignore Zeros. (line 6)
4723 * ignore-zeros, short description: Blocking Factor. (line 156)
4724 * ignore-zeros, summary: Option Summary. (line 305)
4725 * incremental, summary: Option Summary. (line 310)
4726 * incremental, using with --list: Incremental Dumps. (line 177)
4727 * index-file, summary: Option Summary. (line 317)
4728 * info-script: Multi-Volume Archives.
4730 * info-script, short description: Device. (line 104)
4731 * info-script, summary: Option Summary. (line 320)
4732 * interactive: interactive. (line 14)
4733 * interactive, summary: Option Summary. (line 328)
4734 * keep-newer-files: Keep Newer Files. (line 6)
4735 * keep-newer-files, summary: Option Summary. (line 335)
4736 * keep-old-files: Keep Old Files. (line 6)
4737 * keep-old-files, introduced: Dealing with Old Files.
4739 * keep-old-files, summary: Option Summary. (line 339)
4740 * label: label. (line 8)
4741 * label, summary: Option Summary. (line 344)
4742 * list: list. (line 6)
4743 * list, summary: Operation Summary. (line 46)
4744 * list, using with --incremental: Incremental Dumps. (line 177)
4745 * list, using with --listed-incremental: Incremental Dumps. (line 177)
4746 * list, using with --verbose: list. (line 30)
4747 * list, using with file name arguments: list. (line 68)
4748 * listed-incremental: Incremental Dumps. (line 14)
4749 * listed-incremental, summary: Option Summary. (line 351)
4750 * listed-incremental, using with --extract: Incremental Dumps.
4752 * listed-incremental, using with --list: Incremental Dumps. (line 177)
4753 * lzma: gzip. (line 134)
4754 * lzma, summary: Option Summary. (line 359)
4755 * lzop: gzip. (line 138)
4756 * mode: override. (line 14)
4757 * mode, summary: Option Summary. (line 368)
4758 * mtime: override. (line 29)
4759 * mtime, summary: Option Summary. (line 374)
4760 * multi-volume: Multi-Volume Archives.
4762 * multi-volume, short description: Device. (line 88)
4763 * multi-volume, summary: Option Summary. (line 383)
4764 * new-volume-script: Multi-Volume Archives.
4766 * new-volume-script, short description: Device. (line 104)
4767 * new-volume-script, summary: Option Summary. (line 320)
4768 * newer: after. (line 26)
4769 * newer, summary: Option Summary. (line 391)
4770 * newer-mtime: after. (line 37)
4771 * newer-mtime, summary: Option Summary. (line 399)
4772 * no-anchored: controlling pattern-matching.
4774 * no-anchored, summary: Option Summary. (line 404)
4775 * no-auto-compress, summary: Option Summary. (line 408)
4776 * no-check-device, described: Incremental Dumps. (line 95)
4777 * no-check-device, summary: Option Summary. (line 412)
4778 * no-delay-directory-restore: Directory Modification Times and Permissions.
4780 * no-delay-directory-restore, summary: Option Summary. (line 417)
4781 * no-ignore-case: controlling pattern-matching.
4783 * no-ignore-case, summary: Option Summary. (line 423)
4784 * no-ignore-command-error: Writing to an External Program.
4786 * no-ignore-command-error, summary: Option Summary. (line 426)
4787 * no-null, described: nul. (line 15)
4788 * no-null, summary: Option Summary. (line 430)
4789 * no-overwrite-dir, summary: Option Summary. (line 435)
4790 * no-quote-chars, summary: Option Summary. (line 439)
4791 * no-recursion: recurse. (line 13)
4792 * no-recursion, summary: Option Summary. (line 444)
4793 * no-same-owner: Attributes. (line 67)
4794 * no-same-owner, summary: Option Summary. (line 448)
4795 * no-same-permissions, summary: Option Summary. (line 454)
4796 * no-unquote: Selecting Archive Members.
4798 * no-unquote, summary: Option Summary. (line 459)
4799 * no-wildcards: controlling pattern-matching.
4801 * no-wildcards, summary: Option Summary. (line 463)
4802 * no-wildcards-match-slash: controlling pattern-matching.
4804 * no-wildcards-match-slash, summary: Option Summary. (line 466)
4805 * null, described: nul. (line 11)
4806 * null, summary: Option Summary. (line 469)
4807 * numeric-owner: Attributes. (line 73)
4808 * numeric-owner, summary: Option Summary. (line 475)
4809 * occurrence, summary: Option Summary. (line 492)
4810 * old-archive, summary: Option Summary. (line 506)
4811 * one-file-system: one. (line 16)
4812 * one-file-system, summary: Option Summary. (line 509)
4813 * overwrite: Overwrite Old Files. (line 6)
4814 * overwrite, introduced: Dealing with Old Files.
4816 * overwrite, summary: Option Summary. (line 514)
4817 * overwrite-dir: Overwrite Old Files. (line 28)
4818 * overwrite-dir, introduced: Dealing with Old Files.
4820 * overwrite-dir, summary: Option Summary. (line 518)
4821 * owner: override. (line 57)
4822 * owner, summary: Option Summary. (line 522)
4823 * pax-option: PAX keywords. (line 6)
4824 * pax-option, summary: Option Summary. (line 531)
4825 * portability, summary: Option Summary. (line 537)
4826 * posix, summary: Option Summary. (line 541)
4827 * preserve: Attributes. (line 126)
4828 * preserve, summary: Option Summary. (line 544)
4829 * preserve-order: Same Order. (line 6)
4830 * preserve-order, summary: Option Summary. (line 548)
4831 * preserve-permissions: Setting Access Permissions.
4833 * preserve-permissions, short description: Attributes. (line 113)
4834 * preserve-permissions, summary: Option Summary. (line 551)
4835 * quote-chars, summary: Option Summary. (line 561)
4836 * quoting-style: quoting styles. (line 38)
4837 * quoting-style, summary: Option Summary. (line 565)
4838 * read-full-records <1>: read full records. (line 6)
4839 * read-full-records: Reading. (line 8)
4840 * read-full-records, short description: Blocking Factor. (line 172)
4841 * read-full-records, summary: Option Summary. (line 572)
4842 * record-size, summary: Option Summary. (line 577)
4843 * recursion: recurse. (line 24)
4844 * recursion, summary: Option Summary. (line 581)
4845 * recursive-unlink: Recursive Unlink. (line 6)
4846 * recursive-unlink, summary: Option Summary. (line 585)
4847 * remove-files: remove files. (line 6)
4848 * remove-files, summary: Option Summary. (line 590)
4849 * restrict, summary: Option Summary. (line 594)
4850 * rmt-command, summary: Option Summary. (line 599)
4851 * rsh-command: Device. (line 73)
4852 * rsh-command, summary: Option Summary. (line 603)
4853 * same-order: Same Order. (line 6)
4854 * same-order, summary: Option Summary. (line 607)
4855 * same-owner: Attributes. (line 48)
4856 * same-owner, summary: Option Summary. (line 615)
4857 * same-permissions: Setting Access Permissions.
4859 * same-permissions, short description: Attributes. (line 113)
4860 * same-permissions, summary: Option Summary. (line 551)
4861 * seek, summary: Option Summary. (line 624)
4862 * show-defaults: defaults. (line 6)
4863 * show-defaults, summary: Option Summary. (line 631)
4864 * show-omitted-dirs: verbose. (line 107)
4865 * show-omitted-dirs, summary: Option Summary. (line 640)
4866 * show-stored-names: list. (line 60)
4867 * show-stored-names, summary: Option Summary. (line 644)
4868 * show-transformed-names: transform. (line 45)
4869 * show-transformed-names, summary: Option Summary. (line 644)
4870 * sparse: sparse. (line 22)
4871 * sparse, summary: Option Summary. (line 652)
4872 * sparse-version: sparse. (line 57)
4873 * sparse-version, summary: Option Summary. (line 657)
4874 * starting-file: Starting File. (line 6)
4875 * starting-file, summary: Option Summary. (line 662)
4876 * strip-components: transform. (line 25)
4877 * strip-components, summary: Option Summary. (line 668)
4878 * suffix: backup. (line 68)
4879 * suffix, summary: Option Summary. (line 677)
4880 * tape-length: Multi-Volume Archives.
4882 * tape-length, short description: Device. (line 96)
4883 * tape-length, summary: Option Summary. (line 683)
4884 * test-label: label. (line 37)
4885 * test-label, summary: Option Summary. (line 688)
4886 * to-command: Writing to an External Program.
4888 * to-command, summary: Option Summary. (line 692)
4889 * to-stdout: Writing to Standard Output.
4891 * to-stdout, summary: Option Summary. (line 696)
4892 * totals: verbose. (line 46)
4893 * totals, summary: Option Summary. (line 701)
4894 * touch <1>: Attributes. (line 37)
4895 * touch: Data Modification Times.
4897 * touch, summary: Option Summary. (line 706)
4898 * transform: transform. (line 74)
4899 * transform, summary: Option Summary. (line 712)
4900 * uncompress: gzip. (line 141)
4901 * uncompress, summary: Option Summary. (line 152)
4902 * ungzip: gzip. (line 96)
4903 * ungzip, summary: Option Summary. (line 274)
4904 * unlink-first: Unlink First. (line 6)
4905 * unlink-first, introduced: Dealing with Old Files.
4907 * unlink-first, summary: Option Summary. (line 732)
4908 * unquote: Selecting Archive Members.
4910 * unquote, summary: Option Summary. (line 738)
4911 * update: update. (line 8)
4912 * update, summary: Operation Summary. (line 50)
4913 * usage: help. (line 53)
4914 * use-compress-program: gzip. (line 146)
4915 * use-compress-program, summary: Option Summary. (line 742)
4916 * utc, summary: Option Summary. (line 746)
4917 * verbose: verbose. (line 18)
4918 * verbose, introduced: verbose tutorial. (line 6)
4919 * verbose, summary: Option Summary. (line 750)
4920 * verbose, using with --create: create verbose. (line 6)
4921 * verbose, using with --list: list. (line 30)
4922 * verify, short description: verify. (line 8)
4923 * verify, summary: Option Summary. (line 757)
4924 * verify, using with --create: verify. (line 24)
4925 * version: help. (line 6)
4926 * version, summary: Option Summary. (line 762)
4927 * volno-file: Multi-Volume Archives.
4929 * volno-file, summary: Option Summary. (line 767)
4930 * wildcards: controlling pattern-matching.
4932 * wildcards, summary: Option Summary. (line 772)
4933 * wildcards-match-slash: controlling pattern-matching.
4935 * wildcards-match-slash, summary: Option Summary. (line 776)
4936 * xform: transform. (line 74)
4937 * xform, summary: Option Summary. (line 712)
4940 File: tar.info, Node: Index, Prev: Index of Command Line Options, Up: Top
4948 * abbreviations for months: Calendar date items. (line 38)
4949 * absolute file names: Remote Tape Server. (line 17)
4950 * Adding archives to an archive: concatenate. (line 6)
4951 * Adding files to an Archive: appending files. (line 8)
4952 * ADMINISTRATOR: General-Purpose Variables.
4954 * Age, excluding files by: after. (line 8)
4955 * ago in date strings: Relative items in date strings.
4957 * am in date strings: Time of day items. (line 22)
4958 * Appending files to an Archive: appending files. (line 8)
4959 * Arch, excluding files: exclude. (line 39)
4960 * archive: Definitions. (line 6)
4961 * Archive creation: file. (line 36)
4962 * archive member: Definitions. (line 15)
4963 * Archive Name: file. (line 8)
4964 * Archive, creation of: create. (line 8)
4965 * Archives, Appending files to: appending files. (line 8)
4966 * Archiving Directories: create dir. (line 6)
4967 * archiving files: Top. (line 24)
4968 * ARGP_HELP_FMT, environment variable: Configuring Help Summary.
4970 * authors of get_date: Authors of get_date. (line 6)
4971 * Avoiding recursion in directories: recurse. (line 8)
4972 * backup options: backup. (line 6)
4973 * backup suffix: backup. (line 68)
4974 * BACKUP_DIRS: General-Purpose Variables.
4976 * BACKUP_FILES: General-Purpose Variables.
4978 * BACKUP_HOUR: General-Purpose Variables.
4980 * backups: backup. (line 41)
4981 * Bazaar, excluding files: exclude. (line 39)
4982 * beginning of time, for POSIX: Seconds since the Epoch.
4984 * bell, checkpoint action: checkpoints. (line 65)
4985 * Bellovin, Steven M.: Authors of get_date. (line 6)
4986 * Berets, Jim: Authors of get_date. (line 6)
4987 * Berry, K.: Authors of get_date. (line 14)
4988 * Block number where error occurred: verbose. (line 115)
4989 * BLOCKING: General-Purpose Variables.
4991 * blocking factor: Blocking Factor. (line 194)
4992 * Blocking Factor: Blocking Factor. (line 6)
4993 * Blocks per record: Blocking Factor. (line 6)
4994 * bug reports: Reports. (line 6)
4995 * Bytes per record: Blocking Factor. (line 6)
4996 * bzip2: gzip. (line 6)
4997 * calendar date item: Calendar date items. (line 6)
4998 * case, ignored in dates: General date syntax. (line 64)
4999 * cat vs concatenate: concatenate. (line 63)
5000 * Changing directory mid-stream: directory. (line 6)
5001 * Character class, excluding characters from: wildcards. (line 34)
5002 * checkpoints, defined: checkpoints. (line 6)
5003 * Choosing an archive file: file. (line 8)
5004 * comments, in dates: General date syntax. (line 64)
5005 * compress: gzip. (line 6)
5006 * Compressed archives: gzip. (line 6)
5007 * concatenate vs cat: concatenate. (line 63)
5008 * Concatenating Archives: concatenate. (line 6)
5009 * corrupted archives <1>: gzip. (line 115)
5010 * corrupted archives: Full Dumps. (line 8)
5011 * Creation of the archive: create. (line 8)
5012 * CVS, excluding files: exclude. (line 39)
5013 * Darcs, excluding files: exclude. (line 39)
5014 * DAT blocking: Blocking Factor. (line 204)
5015 * Data Modification time, excluding files by: after. (line 8)
5016 * Data modification times of extracted files: Data Modification Times.
5018 * date format, ISO 8601: Calendar date items. (line 30)
5019 * date input formats: Date input formats. (line 6)
5020 * day in date strings: Relative items in date strings.
5022 * day of week item: Day of week items. (line 6)
5023 * Deleting files from an archive: delete. (line 8)
5024 * Deleting from tape archives: delete. (line 19)
5025 * dereferencing hard links: hard links. (line 8)
5026 * Descending directories, avoiding: recurse. (line 8)
5027 * Device numbers, changing: Fixing Snapshot Files.
5029 * Device numbers, using in incremental backups: Incremental Dumps.
5031 * Directories, Archiving: create dir. (line 6)
5032 * Directories, avoiding recursion: recurse. (line 8)
5033 * Directory, changing mid-stream: directory. (line 6)
5034 * DIRLIST: General-Purpose Variables.
5036 * displacement of dates: Relative items in date strings.
5038 * doc-opt-col: Configuring Help Summary.
5040 * dot, checkpoint action: checkpoints. (line 80)
5041 * Double-checking a write operation: verify. (line 6)
5042 * DUMP_BEGIN: User Hooks. (line 32)
5043 * DUMP_END: User Hooks. (line 36)
5044 * DUMP_REMIND_SCRIPT: General-Purpose Variables.
5046 * dumps, full: Full Dumps. (line 8)
5047 * dup-args: Configuring Help Summary.
5049 * dup-args-note: Configuring Help Summary.
5051 * echo, checkpoint action: checkpoints. (line 25)
5052 * Eggert, Paul: Authors of get_date. (line 6)
5053 * End-of-archive blocks, ignoring: Ignore Zeros. (line 6)
5054 * End-of-archive info script: Multi-Volume Archives.
5056 * entry: Naming tar Archives. (line 11)
5057 * epoch, for POSIX: Seconds since the Epoch.
5059 * Error message, block number of: verbose. (line 125)
5060 * Exabyte blocking: Blocking Factor. (line 204)
5061 * exclude: exclude. (line 14)
5062 * exclude-caches: exclude. (line 84)
5063 * exclude-from: exclude. (line 27)
5064 * exclude-tag: exclude. (line 107)
5065 * Excluding characters from a character class: wildcards. (line 34)
5066 * Excluding file by age: after. (line 8)
5067 * Excluding files by file system: exclude. (line 8)
5068 * Excluding files by name and pattern: exclude. (line 8)
5069 * Exec Mode, genfile: Exec Mode. (line 6)
5070 * exec, checkpoint action: checkpoints. (line 96)
5071 * existing backup method: backup. (line 59)
5072 * exit status: Synopsis. (line 67)
5073 * Extraction: extract. (line 8)
5074 * extraction: Definitions. (line 22)
5075 * FDL, GNU Free Documentation License: GNU Free Documentation License.
5077 * file archival: Top. (line 24)
5078 * File lists separated by NUL characters: Generate Mode. (line 33)
5079 * file name: Definitions. (line 15)
5080 * File Name arguments, alternatives: files. (line 6)
5081 * File name arguments, using --list with: list. (line 68)
5082 * File names, excluding files by: exclude. (line 8)
5083 * File names, terminated by NUL: nul. (line 6)
5084 * File names, using hard links: hard links. (line 8)
5085 * File names, using symbolic links: dereference. (line 6)
5086 * File system boundaries, not crossing: one. (line 6)
5087 * FILELIST: General-Purpose Variables.
5089 * first in date strings: General date syntax. (line 26)
5090 * format 0, snapshot file: Snapshot Files. (line 23)
5091 * format 1, snapshot file: Snapshot Files. (line 47)
5092 * format 2, snapshot file: Snapshot Files. (line 69)
5093 * Format Options: Format Variations. (line 6)
5094 * Format Parameters: Format Variations. (line 6)
5095 * Format, old style: old. (line 6)
5096 * fortnight in date strings: Relative items in date strings.
5098 * free documentation: Free Software Needs Free Documentation.
5100 * full dumps: Full Dumps. (line 8)
5101 * future time stamps: Large or Negative Values.
5103 * general date syntax: General date syntax. (line 6)
5104 * Generate Mode, genfile: Generate Mode. (line 6)
5105 * genfile: Genfile. (line 6)
5106 * genfile, create file: Generate Mode. (line 6)
5107 * genfile, creating sparse files: Generate Mode. (line 55)
5108 * genfile, generate mode: Generate Mode. (line 6)
5109 * genfile, reading a list of file names: Generate Mode. (line 22)
5110 * genfile, seeking to a given offset: Generate Mode. (line 18)
5111 * get_date: Date input formats. (line 6)
5112 * Getting program version number: help. (line 6)
5113 * git, excluding files: exclude. (line 39)
5114 * GNU archive format: gnu. (line 6)
5115 * GNU.sparse.major, extended header variable: PAX 1. (line 14)
5116 * GNU.sparse.map, extended header variable: PAX 0. (line 60)
5117 * GNU.sparse.minor, extended header variable: PAX 1. (line 17)
5118 * GNU.sparse.name, extended header variable: PAX 0. (line 68)
5119 * GNU.sparse.name, extended header variable, in v.1.0: PAX 1. (line 24)
5120 * GNU.sparse.numblocks, extended header variable: PAX 0. (line 15)
5121 * GNU.sparse.numbytes, extended header variable: PAX 0. (line 21)
5122 * GNU.sparse.offset, extended header variable: PAX 0. (line 18)
5123 * GNU.sparse.realsize, extended header variable: PAX 1. (line 24)
5124 * GNU.sparse.size, extended header variable: PAX 0. (line 11)
5125 * gnupg, using with tar: gzip. (line 158)
5126 * gpg, using with tar: gzip. (line 158)
5127 * gzip: gzip. (line 6)
5128 * hard links, dereferencing: hard links. (line 8)
5129 * header-col: Configuring Help Summary.
5131 * hook: User Hooks. (line 13)
5132 * hour in date strings: Relative items in date strings.
5134 * Ignoring end-of-archive blocks: Ignore Zeros. (line 6)
5135 * Info script: Multi-Volume Archives.
5137 * Interactive operation: interactive. (line 6)
5138 * ISO 8601 date format: Calendar date items. (line 30)
5139 * items in date strings: General date syntax. (line 6)
5140 * Labeling an archive: label. (line 6)
5141 * Labeling multi-volume archives: label. (line 6)
5142 * Labels on the archive media: label. (line 6)
5143 * language, in dates: General date syntax. (line 40)
5144 * Large lists of file names on small machines: Same Order. (line 6)
5145 * large values: Large or Negative Values.
5147 * last DAY: Day of week items. (line 15)
5148 * last in date strings: General date syntax. (line 26)
5149 * Listing all tar options: help. (line 26)
5150 * listing member and file names: list. (line 41)
5151 * Listing volume label: label. (line 29)
5152 * Lists of file names: files. (line 6)
5153 * Local and remote archives: file. (line 73)
5154 * long-opt-col: Configuring Help Summary.
5156 * lzma: gzip. (line 6)
5157 * lzop: gzip. (line 6)
5158 * MacKenzie, David: Authors of get_date. (line 6)
5159 * member: Definitions. (line 15)
5160 * member name: Definitions. (line 15)
5161 * Members, replacing with other members: append. (line 49)
5162 * Mercurial, excluding files: exclude. (line 39)
5163 * Meyering, Jim: Authors of get_date. (line 6)
5164 * Middle of the archive, starting in the: Starting File. (line 11)
5165 * midnight in date strings: Time of day items. (line 22)
5166 * minute in date strings: Relative items in date strings.
5168 * minutes, time zone correction by: Time of day items. (line 30)
5169 * Modes of extracted files: Setting Access Permissions.
5171 * Modification time, excluding files by: after. (line 8)
5172 * Modification times of extracted files: Data Modification Times.
5174 * month in date strings: Relative items in date strings.
5176 * month names in date strings: Calendar date items. (line 38)
5177 * months, written-out: General date syntax. (line 36)
5178 * MT: General-Purpose Variables.
5180 * MT_BEGIN: Magnetic Tape Control.
5182 * MT_OFFLINE: Magnetic Tape Control.
5184 * MT_REWIND: Magnetic Tape Control.
5186 * MT_STATUS: Magnetic Tape Control.
5188 * Multi-volume archives: Multi-Volume Archives.
5190 * Mutli-volume archives in PAX format, extracting using non-GNU tars: Split Recovery.
5192 * Mutli-volume archives, extracting using non-GNU tars: Split Recovery.
5194 * Naming an archive: file. (line 8)
5195 * negative time stamps: Large or Negative Values.
5197 * next DAY: Day of week items. (line 15)
5198 * next in date strings: General date syntax. (line 26)
5199 * noon in date strings: Time of day items. (line 22)
5200 * now in date strings: Relative items in date strings.
5202 * ntape device: Many. (line 6)
5203 * NUL terminated file names: nul. (line 6)
5204 * Number of blocks per record: Blocking Factor. (line 6)
5205 * Number of bytes per record: Blocking Factor. (line 6)
5206 * numbered backup method: backup. (line 55)
5207 * numbers, written-out: General date syntax. (line 26)
5208 * Obtaining help: help. (line 26)
5209 * Obtaining total status information: verbose. (line 46)
5210 * Old GNU archive format: gnu. (line 6)
5211 * Old GNU sparse format: Old GNU Format. (line 6)
5212 * Old style archives: old. (line 6)
5213 * Old style format: old. (line 6)
5214 * opt-doc-col: Configuring Help Summary.
5216 * option syntax, traditional: Old Options. (line 60)
5217 * Options when reading archives: Reading. (line 6)
5218 * Options, archive format specifying: Format Variations. (line 6)
5219 * Options, format specifying: Format Variations. (line 6)
5220 * ordinal numbers: General date syntax. (line 26)
5221 * Overwriting old files, prevention: Dealing with Old Files.
5223 * pattern, genfile: Generate Mode. (line 39)
5224 * PAX archive format: posix. (line 6)
5225 * Permissions of extracted files: Setting Access Permissions.
5227 * Pinard, F.: Authors of get_date. (line 14)
5228 * pm in date strings: Time of day items. (line 22)
5229 * POSIX archive format: posix. (line 6)
5230 * Progress information: verbose. (line 83)
5231 * Protecting old files: Dealing with Old Files.
5233 * pure numbers in date strings: Pure numbers in date strings.
5235 * RCS, excluding files: exclude. (line 39)
5236 * Reading file names from a file: files. (line 6)
5237 * Reading incomplete records: Reading. (line 8)
5238 * Record Size: Blocking Factor. (line 6)
5239 * Records, incomplete: Reading. (line 8)
5240 * Recursion in directories, avoiding: recurse. (line 8)
5241 * relative items in date strings: Relative items in date strings.
5243 * Remote devices: file. (line 62)
5244 * remote tape drive: Remote Tape Server. (line 6)
5245 * Removing files from an archive: delete. (line 8)
5246 * Replacing members with other members: append. (line 49)
5247 * reporting bugs: Reports. (line 6)
5248 * RESTORE_BEGIN: User Hooks. (line 39)
5249 * RESTORE_END: User Hooks. (line 42)
5250 * Resurrecting files from an archive: extract. (line 8)
5251 * Retrieving files from an archive: extract. (line 8)
5252 * return status: Synopsis. (line 67)
5253 * rmargin: Configuring Help Summary.
5255 * rmt: Remote Tape Server. (line 6)
5256 * RSH: General-Purpose Variables.
5258 * RSH_COMMAND: General-Purpose Variables.
5260 * Running out of space: Scarce. (line 8)
5261 * Salz, Rich: Authors of get_date. (line 6)
5262 * SCCS, excluding files: exclude. (line 39)
5263 * short-opt-col: Configuring Help Summary.
5265 * simple backup method: backup. (line 64)
5266 * SIMPLE_BACKUP_SUFFIX: backup. (line 68)
5267 * sleep, checkpoint action: checkpoints. (line 90)
5268 * SLEEP_MESSAGE: General-Purpose Variables.
5270 * SLEEP_TIME: General-Purpose Variables.
5272 * Small memory: Scarce. (line 8)
5273 * snapshot file, format 0: Snapshot Files. (line 23)
5274 * snapshot file, format 1: Snapshot Files. (line 47)
5275 * snapshot file, format 2: Snapshot Files. (line 69)
5276 * snapshot files, editing: Fixing Snapshot Files.
5278 * snapshot files, fixing device numbers: Fixing Snapshot Files.
5280 * Sparse Files: sparse. (line 6)
5281 * sparse files v.0.0, extracting with non-GNU tars: Sparse Recovery.
5283 * sparse files v.0.1, extracting with non-GNU tars: Sparse Recovery.
5285 * sparse files v.1.0, extracting with non-GNU tars: Sparse Recovery.
5287 * Sparse files, creating using genfile: Generate Mode. (line 55)
5288 * sparse files, extracting with non-GNU tars: Sparse Recovery.
5290 * sparse formats: Sparse Formats. (line 6)
5291 * sparse formats, defined: sparse. (line 50)
5292 * sparse formats, Old GNU: Old GNU Format. (line 6)
5293 * sparse formats, v.0.0: PAX 0. (line 6)
5294 * sparse formats, v.0.1: PAX 0. (line 52)
5295 * sparse formats, v.1.0: PAX 1. (line 6)
5296 * sparse versions: Sparse Formats. (line 6)
5297 * Specifying archive members: Selecting Archive Members.
5299 * Specifying files to act on: Selecting Archive Members.
5301 * Standard input and output: file. (line 41)
5302 * Standard output, writing extracted files to: Writing to Standard Output.
5304 * Storing archives in compressed format: gzip. (line 6)
5305 * SVN, excluding files: exclude. (line 39)
5306 * Symbolic link as file name: dereference. (line 6)
5307 * TAPE: file tutorial. (line 14)
5308 * tape blocking: Blocking Factor. (line 194)
5309 * tape marks: Many. (line 44)
5310 * tape positioning: Many. (line 26)
5311 * TAPE_FILE: General-Purpose Variables.
5313 * Tapes, using --delete and: delete. (line 19)
5314 * TAR: General-Purpose Variables.
5316 * tar: What tar Does. (line 6)
5317 * tar archive: Definitions. (line 6)
5318 * Tar archive formats: Formats. (line 6)
5319 * tar entry: Naming tar Archives. (line 11)
5320 * tar file: Naming tar Archives. (line 11)
5321 * tar to a remote device: file. (line 62)
5322 * tar to standard input and output: file. (line 41)
5323 * tar-snapshot-edit: Fixing Snapshot Files.
5325 * TAR_ARCHIVE, checkpoint script environment: checkpoints. (line 108)
5326 * TAR_ARCHIVE, info script environment variable: Multi-Volume Archives.
5328 * TAR_ATIME, to-command environment: Writing to an External Program.
5330 * TAR_BLOCKING_FACTOR, checkpoint script environment: checkpoints.
5332 * TAR_BLOCKING_FACTOR, info script environment variable: Multi-Volume Archives.
5334 * TAR_CHECKPOINT, checkpoint script environment: checkpoints. (line 114)
5335 * TAR_CTIME, to-command environment: Writing to an External Program.
5337 * TAR_FD, info script environment variable: Multi-Volume Archives.
5339 * TAR_FILENAME, to-command environment: Writing to an External Program.
5341 * TAR_FILETYPE, to-command environment: Writing to an External Program.
5343 * TAR_FORMAT, checkpoint script environment: checkpoints. (line 121)
5344 * TAR_FORMAT, info script environment variable: Multi-Volume Archives.
5346 * TAR_GID, to-command environment: Writing to an External Program.
5348 * TAR_GNAME, to-command environment: Writing to an External Program.
5350 * TAR_MODE, to-command environment: Writing to an External Program.
5352 * TAR_MTIME, to-command environment: Writing to an External Program.
5354 * TAR_OPTIONS, environment variable: using tar options. (line 30)
5355 * TAR_REALNAME, to-command environment: Writing to an External Program.
5357 * TAR_SIZE, to-command environment: Writing to an External Program.
5359 * TAR_SUBCOMMAND, checkpoint script environment: checkpoints. (line 117)
5360 * TAR_SUBCOMMAND, info script environment variable: Multi-Volume Archives.
5362 * TAR_UID, to-command environment: Writing to an External Program.
5364 * TAR_UNAME, to-command environment: Writing to an External Program.
5366 * TAR_VERSION, checkpoint script environment: checkpoints. (line 105)
5367 * TAR_VERSION, info script environment variable: Multi-Volume Archives.
5369 * TAR_VOLUME, info script environment variable: Multi-Volume Archives.
5371 * tarcat: Tarcat. (line 6)
5372 * this in date strings: Relative items in date strings.
5374 * time of day item: Time of day items. (line 6)
5375 * time zone correction: Time of day items. (line 30)
5376 * time zone item <1>: Time zone items. (line 6)
5377 * time zone item: General date syntax. (line 44)
5378 * today in date strings: Relative items in date strings.
5380 * tomorrow in date strings: Relative items in date strings.
5382 * ttyout, checkpoint action: checkpoints. (line 70)
5383 * TZ: Specifying time zone rules.
5385 * Ultrix 3.1 and write failure: Remote Tape Server. (line 40)
5386 * unpacking: Definitions. (line 22)
5387 * Updating an archive: update. (line 8)
5388 * usage-indent: Configuring Help Summary.
5390 * Using encrypted archives: gzip. (line 158)
5391 * ustar archive format: ustar. (line 6)
5392 * uuencode: Applications. (line 8)
5393 * v7 archive format: old. (line 6)
5394 * VCS, excluding files: exclude. (line 39)
5395 * Verbose operation: verbose. (line 18)
5396 * Verifying a write operation: verify. (line 6)
5397 * Verifying the currency of an archive: compare. (line 6)
5398 * version control system, excluding files: exclude. (line 39)
5399 * Version of the tar program: help. (line 6)
5400 * version-control Emacs variable: backup. (line 49)
5401 * VERSION_CONTROL: backup. (line 41)
5402 * volno file: Multi-Volume Archives.
5404 * VOLNO_FILE: General-Purpose Variables.
5406 * Volume label, listing: label. (line 29)
5407 * Volume number file: Multi-Volume Archives.
5409 * week in date strings: Relative items in date strings.
5411 * Where is the archive?: file. (line 8)
5412 * Working directory, specifying: directory. (line 6)
5413 * Writing extracted files to standard output: Writing to Standard Output.
5415 * Writing new archives: file. (line 36)
5416 * XLIST: General-Purpose Variables.
5418 * xsparse: Sparse Recovery. (line 13)
5419 * year in date strings: Relative items in date strings.
5421 * yesterday in date strings: Relative items in date strings.